Effects of Two Melt Extrusion Based Additive Manufacturing Technologies and Common Sterilization Methods on the Properties of a Medical Grade PLGA Copolymer

Abstract

Although bioabsorbable polymers have garnered increasing attention because of their potential in tissue engineering applications, to our knowledge there are only a few bioabsorbable 3D printed medical devices on the market thus far. In this study, we assessed the processability of medical grade Poly(lactic-co-glycolic) Acid (PLGA)85:15 via two additive manufacturing technologies: Fused Filament Fabrication (FFF) and Direct Pellet Printing (DPP) to highlight the least destructive technology towards PLGA. To quantify PLGA degradation, its molecular weight (gel permeation chromatography (GPC)) as well as its thermal properties (differential scanning calorimetry (DSC)) were evaluated at each processing step, including sterilization with conventional methods (ethylene oxide, gamma, and beta irradiation). Results show that 3D printing of PLGA on a DPP printer significantly decreased the number-average molecular weight (M_n) to the greatest extent (26% M_n loss, p < 0.0001) as it applies a longer residence time and higher shear stress compared to classic FFF (19% M_n loss, p < 0.0001). Among all sterilization methods tested, ethylene oxide seems to be the most appropriate, as it leads to no significant changes in PLGA properties. After sterilization, all samples were considered to be non-toxic, as cell viability was above 70% compared to the control, indicating that this manufacturing route could be used for the development of bioabsorbable medical devices. Based on our observations, we recommend using FFF printing and ethylene oxide sterilization to produce PLGA medical devices.

Keywords: additive manufacturing; bioabsorbable; medical devices; polymer; sterilization.

Figure 1

(a) Description of the manufacturing process n°1, including extrusion of the Poly(Lactide-co-Glycolide) (PLGA) filament and Fused Filament Fabrication; (b) Description of the manufacturing process n°2: Direct Pellet Printing; (c) Dimensions of the 3D printed sample.

Figure 2

(a) Comparison of mass of PLGA samples printed on FFF and PAM printer. All data were analyzed using a non parametric student t-test *** p < 0.001, (b) FFF and (c) DPP printed samples.

Figure 3

(a) Evaluation of (a) the number-average molecular weight (M_n), (b) weight-average (M_w), and (c) polydispersity index (IP) of PLGA 85:15 from successively printed samples; All data are expressed as mean ± SD (n = 3).

Figure 4

(a) Impact of additive manufacturing steps on the (a) M_n, (b) M_w and (c) polydispersity index of PLGA 85:15; All data are expressed as mean ± SD (n = 3) and were analyzed using Anova one-way test and compared to the control * p < 0.05, *** p < 0.001, **** p < 0.0001.

Figure 5

(a) Differential scanning calorimetry (DSC) thermograms of PLGA 85:15 raw material, filament, PAM, and Ultimaker 3 3D printed constructs with a heating rate of 10 °C/min for the first heating run. (b) TGA thermogram of PLGA 85:15 pellets.

Figure 6

Impact of ethylene oxide as well as beta, rand, and gamma irradiation sterilization methods on the (a) M_n, (b) M_w, and (c) polydispersity index of PLGA 85:15; All data are expressed as mean ± SD (n = 3) and were analyzed using Anova one-way test and compared to the control * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Figure 7

Impact of ethylene oxide, as well as beta, rand, and gamma irradiation sterilization methods on the (a) Glass transition temperature (T_g), and (b) the crystallinity of PLGA 85:15; All data are expressed as mean ± SD (n = 3) and were analyzed using Anova one-way test and compared to control * p < 0.05, ** p < 0.01.

Figure 8

Impact of ethylene oxide, as well as beta, rand, and gamma irradiation methods on cytotoxicity of PLGA constructs after sterilization; All data are expressed as mean ± SD (n = 8).