SUPPLEMENTARY INFORMATION

Transcription

1 VOLUME: 1 ARTICLE NUMBER: 8 In the format provided by the authors and unedited. A Fully Functional Drug-Eluting Joint Implant 1,2,3 Suhardi VJ, 1,2 Bichara DA, Kwok SJJ 3,4, Freiberg AA 2, Rubash H 2, Malchau H 2, Yun SH 3,4, 1,2 Muratoglu OK, 1,2 Oral E * 1 Harris Orthopaedic Laboratory, Massachusetts General Hospital, Boston, MA. 2 Department of Orthopaedic Surgery, Harvard Medical School. 3 Department of Medical Engineering and Medical Physics, Massachusetts Institute of Technology. 4 Wellmann Center for Photomedicine, Massachusetts General Hospital, Boston, MA *Corresponding Author: Ebru Oral Address: 55 Fruit Street, GRJ 126, Boston, MA 2215, USA Phone: Fax: eoral@partners.org NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

2 Table of Content Supplementary Discussion... 3 Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Figure Supplementary Table Supplementary Table Supplementary Table Supplementary Table NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

3 Supplementary Discussion 1 Strength and Wear Resistance Requirements of UHMWPE for Load-Bearing Prosthetic Joints The mechanical and tribological requirements of UHMWPE as load-bearing prosthetic joints are dictated by the biomechanics of each joint. In the hip, UHMWPE is used as acetabular cup with maximum contact stress of MPa. 1 In the knee, UHMWPE is used as tibial insert with higher maximum contact stress of MPa. 2,3 In addition to strength requirements to resist deformation and fracture against contact stresses, wear resistance is required against multidirectional articulating motion between UHMWPE and its counter surface in the joints. Multidirectional, but not unidirectional motion has been shown to induce transverse rupture of elongated UHMWPE fibrils, resulting in the wear of UHMWPE bearing surfaces. 4 Wear resistance is an essential requirement because wear particles have long been associated with periprosthetic osteolysis, which leads to loosening and failure of the implants. Radiation cross-linking is the universally accepted method of increasing wear resistance of UHMWPE bearing surfaces. Increasing radiation dose is used to increase wear resistance but also decreases mechanical strength and toughness. The lowest radiation dose used is for conventional UHMWPE, which receives radiation only for the purpose of sterilization is in the (-4 kgy) range. There are number of highly cross-linked UHMWPEs (>4 kgy irradiated), which have been developed for higher wear resistance. 5 In total hips, where wear resistance requirements are higher (due to higher frequency of multidirectional motion), about 95% of all joints comprise a highly cross-linked UHMWPE bearing surface 6 in contrast to about 5% of total knee replacements because wear resistance requirements are less stringent (due to higher frequency of unidirectional motion). Tensile mechanical testing, impact testing and in vitro pin-on-disc wear testing or simulator testing are the most common methods of evaluating UHMWPE formulations in vitro. 5 Conventional UHMWPE has an ultimate tensile strength (UTS) of 47-5 MPa 7, an elongation to break (EAB) of % 7, impact strength (IS) of 9-96 kj/mm 2, 8 and wear rate of 6-11 mg/million cycle. 9 Highly-crosslinked UHMWPEs have UTS of MPa 7, EAB of % 11, IS of kj/mm 2,8, and wear rates of mg/million cycle 9. Highly cross-linked UHMWPEs (without distinction of dose at this point) have decreased the incidence of periprosthetic osteolysis 87% over the last decade 1 compared to conventional UHMWPE in total hips. Our goal here was to develop methods by which therapeutic agents are incorporated into UHMWPE bearing surfaces but not increase any risk associated with their use while adding the benefit of antibacterial properties. However, it is also expected that any changes to the structure of the polymer may result in the compromise of one or more properties. There are UHMWPEs 1 Unpublished data for E1 Biomet NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

4 with a range of properties that are successfully used in different applications and in different patient populations according to the clinicians discretion as briefly explained above. Drug eluting UHMWPE with mechanical and tribological properties within the limits of conventional and highly-crosslinked UHMWPEs are expected to perform well as part of fully weight bearing prosthetic joint. References 1. Hua, X. et al. Experimental validation of finite element modelling of a modular metal-onpolyethylene total hip replacement. Proc. Inst. Mech. Eng. H 228, (214). 2. Van Den Heever, D. J., Scheffer, C., Erasmus, P. J. & Dillon, E. M. Contact stresses in a patientspecific unicompartmental knee replacement. Clin. Biomech. 26, (211). 3. Villa, T., Migilavacca, F., Gastaldi, D., Colombo, M. & Pietrabissa, R. Contact stresses and fatigue life in a knee prosthesis: comparison between in vitro measurements and computational simulation. J. Biomech. 18, (24). 4. Wang, A. et al. Orientation softening in the deformation and wear of ultra-high polyethylene. Wear 23 24, (1997). 5. Kurtz, S. M. in UHMWPE Biomaterials Handbook, 3rd edn (Ed. Kurtz, S. M.) (Elsevier, (216). 6. Third AJRR Annual Report on Hip and Knee Arthroplasty Data (American Joint Replacement Registry, 216). 7. Pruitt, L. A. in Total Knee Arthroplasty: A Guide to Get Better Performance (eds Bellemans, J. et al.) (Springer, 25). 8. Oral, E. et al. A surface crosslinked UHMWPE stabilized by vitamin E with low wear and high fatigue strength. Biomaterials 31, (21). 9. Baykal, D., et al. Advances in tribological testing of artificial joint biomaterials using multidirectional pin-on-disk testers. J. Mech. Behav. Biomed. 31, (214) Annual Report on National Joint Replacement Registry. Fig KT29 (Australian Orthopaedic Association, 215) NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

5 Supplemental Figures Supplementary Figure 1. SEM Micrograph and 3D reconstructed μ-ct of highlyeccentric, vancomycin eluting UHMWPE at 2 wt%, 4 wt %, 6 wt %, and 1 wt % initial drug content. Scale bar = 2 μm Porosity (%) Initial drug content (wt %) (a) Accessible Pore (%) Initial drug content (wt %) (b) Supplementary Figure 2. (a) Relation between porosity and initial drug content (wt %). Data are shown as mean ± sd. (b) Relation between accessible pore (%) from the face of the material and initial drug content (wt %). Data are shown as mean ± sd, n=6. NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

6 Area of Inhibition (mm 2 ) S. aureus Days (a) VPE BC Area of Inhibition (mm 2 ) S. Epidermidis VPE BC Days (b) Supplementary Figure 3. Kirby-Bauer agar diffusion test of BC and V-PE against S aureus (a) and S epidermidis (b). Empty area around the samples where there was no visible bacteria growth is called area of inhibition. Data is shown as mean±s.d. (n=5). NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

7 Elution Rate (ug/ml/hr) VPE (Vancomycin) PBS SF Time (hr) (a) Elution Rate (ug/ml/hr) BC (Vancomycin) PBS SF Time (hr) (b) Elution Rate (ug/ml/hr) RVPE (Vancomycin) SF PBS Time (hr) (c) Elution Rate (ug/ml/hr) RVPE (Rifampin) PBS SF Time (hr) Supplementary Figure 4. Antibiotic elution from VPE (vancomycin) (a), vancomycin-bone cement (BC) (vancomycin) (b), and RVPE (vancomycin) (c), and RVPE (Rifampin) (d) in synovial fluid (SF) and phosphate buffered saline (PBS). (d) NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

8 Supplementary Figure 5. Elongation to break (%) of VPE and RVPE. Blue line represents the minimum known yield strength of clinically used UHMWPE as reported by Collier et al, 23 [29]. Green line represents the mean value from 11 % vancomycin bone cement. Supplementary Figure 6. Yield strength (MPa) of VPE and RVPE. Blue line represents the minimum known yield strength of clinically used UHMWPE as reported by Collier et al, 23 [29]. NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

9 Supplementary Figure 7. Wear Rate (mg/mc) of VPE and RVPE. Blue line represents the minimum known wear rate of 25 kgy gamma-irradiated UHMWPE (also called conventional UHMWPE). NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

10 Supplementary Figure 8. SEM Micrograph of various drugs incorporated in to ultra-high molecular weight polyethylene at 4 wt % initial drug concentration. Supplementary Figure 9. Relation between drug polar surface area (PSA) / molecular volume (V) and log of drug elution rate. More polar compounds (PSA/MV >.3) had a higher elution rate at earlier time points compared to non-polar compounds (PSA/MV <.3). As elution times progressed, the elution rate of more polar drugs dropped more rapidly than the non-polar counterparts. NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

11 (a) UTS (MPa) (c) Wear Rate (mg/mc) Irradiation Dose (kgy) 25 1 Irradiation Dose (kgy) (b) (d) Elution Rate (ug/ml/hr) Impact Strength (kj/m 2 ) Irradiation Dose (kgy) kgy 25 kgy 1 kgy Time (hr) Supplementary Figure 1. Ultimate tensile strength (UTS, a), Impact strength (b), and wear rate (c) of VPE receiving various electron beam irradiation doses ( kgy, 25 kgy, and 1 kgy). Data are derived from n=5 for UTS and impact strength. Data are deived from n=3 for wear rate. Data are displayed as means±s.d. NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

12 (a) (b).8.6 OD Positive Control kgy 25 kgy 1 kgy Negative Control Supplementary Figure 11. Long-term antibacterial performance of VPE and RVPE. (a) Planktonic bacteria in the media after 24 hr incubation with V-PE that had been pre-eluted for 6 months and RV-PE that had been pre-eluted for 12 months. Negative control is fresh media, positive control is media with bacteria that had also been incubated for 24 hr. Data is displayed as mean ± s.d. (b) Immunofluorescence staining of grafted vancomycin on the surface of control UHMWPE, V-PE that was irradiated,25, and 1 kgy and had been preeluted for 6 months. Scale bars = 2 um NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

13 Control- Planktonic Supplementary Figure 12. Post-Mortem bioluminescent imaging of rabbit knee in the planktonic control group OD Control VTBC VPE Supplementary Figure 13. Sonication and reculturing of femur, tibia, meniscus, patella, and explants from control, VTBC, and VPE rabbits. Presence of bacteria in the media after sonication and reculturing for 24 hr was measured as absorbtion at 6 nm (OD6). NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

14 (a) 2.5 (b) 4 Creatinine (mg/dl) BUN (mg/dl) Pre-op POD 3 POD 7 POD 21 Pre-op POD 3 POD 7 POD 21 (c) 15 (d) 1 ALT (IU/L) 1 5 ALP (IU/L) Pre-op POD 3 POD 7 POD 21 Pre-op POD 3 POD 7 POD 21 Supplementary Figure 14. Serum creatinine (a), Blood Urea Nitrogen (BUN) (b), serum alanine transaminase (ALT) (c), serum alkaline phosphatase (ALP) (d) of rabbits treated receiving V-PE plugs. Bloods were drawn and analyzed prior to surgery (Pre-op), post operative day 3 (POD 3), post- operative day 7 (POD 7), and post-operative day 21 (POD 21). Data are displayed as mean ± s.d. NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

15 Vanco/Rifampin Release Rate :1 1.5:1 2.:1 2.5: Time (hr) Supplementary Figure 15. Ratio of eluted rifampin and vancomycin throughout elution time at different initial drugs loading ratio. Data are displayed as mean ± s.d., n=6 Control- Biofilm Supplementary Figure 16. Post-Mortem bioluminescent imaging of rabbit knee in the biofilm control group. NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

16 (a) (b).8 Total Flux (p/s) OD Control VTBC RVPE. Control VTBC RVPE Supplementary Figure 17. Total bioluminescence flux (a) and absorption at 6 nm (OD6) (b), of rabbits in biofilm bacterial study receiving control UHMWPE plugs (control), antibiotic eluting bone cement (VTBC), or RVPE plugs (RVPE) Creatinine (mg/dl) BUN (mg/dl) Pre-op POD 3 POD 7 POD 21 5 Pre-op POD 3 POD 7 POD 21 ALT (IU/L) 1 5 ALP (IU/L) Pre-op POD 3 POD 7 POD 21 Pre-op POD 3 POD 7 POD 21 Supplementary Figure 18. Serum creatinine (a), Blood Urea Nitrogen (BUN) (b), serum alanine transaminase (ALT) (c), serum alkaline phosphatase (ALP) (d) of rabbits treated receiving RV- PE plugs. Bloods were drawn and analyzed prior to surgery (Pre-op), post operative day 3 (POD 3), post-operative day 7 (POD 7), and post operative day 21 (POD 21). Data are displayed as mean ± s.d., n=5 NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

17 (a) (b) (c).6 BV/TV.4.2. Control RVPE Supplementary Figure 19. Effect of RVPE on bony ongrowth in murine model. (a) Schematic and representative gross view of implantation of conventional-pe (control), RVPE, and stainless steel screw. (b) Representative micro-ct image of rat tibia receiving stainless screw after six weeks since implantation. The control UHMWPE (n=4) or RVPE (n=4) plugs were implanted transcondylarly on distal femur. Orange indicate hard tissue, yellow indicate soft tissue and empty space, red indicate screw. (c) Quantification of bone volume/total volume (BV/TV) of the bone surrounding the screws. No statistical significance difference between control and RVPE was observed. NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

18 Supplementary Table 1. Calculation of RV-PE to Clinically Relevant Vancomycin to Rifampin Ratio Parameter Magnitude Unit Vancomycin Dose (A) 1 15 mg/kg per 12 hr Rifampin Dose (B) 1 45 mg/kg per 12 hr Serum trough concentration of rifampin (C) ug/ml Serum trough concentration of vancomycin (D) 3 2 ug/ml Rifampin penetration to bone(e) 4 1 % Vancomycin penetration to bone (F) 4 21 % Rifampin concentration in infected bone (G=C*E) Vancomycin Concentration in infected bone (H=D*F) 4.2 Vancomycin to Rifampin Ratio (H/G) Osmon, D.R., et al. Diagnosis and Management of Prosthetic Joint Infection: Clinical Practice Guidelines by the Infectious Diseases Society of America, Clin Infect Diseases, 213, 56, e1 25 (213). 2. Garnham, J. C., Taylor, T., Turner, P., Chasseaud, L.F. Serum concentrations and bioavailability of rifampicin and isoniazid in combination Br. J. clin. Pharmac, 1976, 3, Liu, C., et al. Clinical Practice Guidelines by the Infectious Diseases Society of America for the Treatment of Methicillin Resistant Staphylococcus Aureus Infections in Adults and Children. Clin. Infect. Dis., 52, 1 38 (211). 4. Spellberg, B., Lipsky, B.A. Diagnosis and Management of Prosthetic Joint Infection: Clinical Practice Guidelines by the Infectious Diseases Society of America, Clin. Infect. Dis., 54, (212). NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

19 Supplementary Table 2. Calculation of RV-PE to Titanium-Bone interface Ratio Parameter Magnitude Unit In Vivo Synovial fluid volume in the knee 1 3. ± 1.1 ml Approximate tibial insert dimension ( L x W x H) 4 x 8 x 6 mm Titanium-Bone interface surface area from tibial tray (approximate using 4 x 8 mm as contact area) Titanium-Bone interface surface area from tibial tray (approximated to be twice the surface area from tibial tray) 3.2 x 1 3 mm x 1 3 mm 2 Total titanium-bone interface area 9.6 x 1 3 mm 2 In Vitro Titanium-Bone Interface Area 7.1 mm 2 RV-PE surface area 2.4 mm 2 Media * 2.2 μl 1 Kraus, V.B., Stabler, T. V., Kong, S.Y., Varju, G, McDaniel, G. Measurement of synovial fluid volume using urea. Osteoarthritis Cartilage, 15, (27). * 1 ml media was used instead of 2.2 ul to ensure the bone, titanium, and RV-PE were all immersed. Using 1 ml instead of 2.2 ul was a worse case scenario because the eluted antibiotics were diluted by ~45 times. NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

20 Supplementary Table 3. Calculation of RV-PE to Screw-Bone interface Ratio for Murine Study Parameter Magnitude Unit In Vivo Approximate tibial insert dimension ( L x W x H) 4 x 8 x 6 mm Approximate tibial insert surface area containing RVPE (A) 344 mm 2 Titanium-Bone interface surface area from tibial tray (approximate using 4 x 8 mm as contact area) Titanium-Bone interface surface area from tibial tray (approximated to be twice the surface area from tibial tray) 3.2 x 1 3 mm x 1 3 mm 2 Total titanium-bone interface area (B) 9.6 x 1 3 mm 2 B/A 2.8 Murine Knee Screw-Bone Interface Area 39.4 mm 2 RV-PE surface area 14.2 mm 2 Ratio 2.8 NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s

21 Supplementary Table 4. Implant Design Based on VPE and RVPE and Comparison of Their Mechanical and Wear Rate to Conventional UHMWPE, Highly-Crosslinked UHMWPE (HXLPE), and ASTM F648 Implant UTS (Mpa) EAB (%) Impact Strength (kj/mm 2 ) Wear Rate (mg/mc) VPE (Brown = VPE) 32.9 ± ± ± ±1.3 RVPE (Red =RVPE, White = Highly-Crosslinked UHMWPE) 31.4 ±5.3 (RVPE) 228 ±22 (RVPE) 78.7 ±3.9 (RVPE) 9. ±1. (RVPE) 34. ±3. (HXLPE) 23 ±17 (HXLPE) 62.2 ±1.6 (HXLPE).6 ±. (HXLPE) VPE+RVPE (Red=RVPE, Brown = VPE) 31.4 ±5.3 (RVPE) 228 ±22 (RVPE) 78.7 ±3.9 (RVPE) 9. ±1. (RVPE) 32.9 ±1.9 (VPE) 245 ±22 (VPE) 79.5 ±3.4 (VPE) 9.6 ±1.3 (VPE) Conventional UHMWPE 46.2± ± ± ±.8 Highly-Crosslinked UHMWPE (HXLPE) 34.±3. 23 ± ± ±.3 ASTM F N/A NATURE BIOMEDICAL ENGINEERING DOI: 1.138/s