Retinal degenerative diseases, such as retinitis pigmentosa, represent a significant fraction of the burden of blindness. This kind of diseases involves the activation of microglial cells which causes neuroinflammation. Flavonoids, such as naringenin (Nar), are able to modulate the pro-inflammatory behavior of the microglia cells but their low bioavailability often hinders this activity. In this context, the aim of this research activity was the development of implantable drug delivery systems through Nar encapsulation into a polymeric matrix, as a valuable way to increase its bioavailability. For this purpose, a blend between microbial poly(3-hydroxybutyrate-co-3-hydroxyvaerate) (PHBV) and poly(d,l-lactide-co-glycolide (PLGA) was processed by means of two different fabrication techniques, i.e., melt-spinning and wet-spinning, to fabricate rods designed for ocular implantation. The two developed kinds of rod had different morphology and thermal properties, as demonstrated by scanning electron microscopy, thermogravimetric analysis, and differential scanning calorimetry. In addition, an increase in Nar concentration in the starting polymeric mixture (1–5 wt% respect to device weight) determined a higher drug encapsulation efficiency, with a resulting drug loading increasing from 0.5 to 3.5 % and from 0.4 to 4.4 % for melt- and wet-spun rods, respectively. A significant effect of processing technique and drug loading on the resulting release kinetics was also observed, with the devices loaded with a higher percentage of Nar reaching a 100 % cumulative drug release after 38 and 73 days in the case of wet- and melt-spun rods, respectively. In addition, the wet-spun rods showed a first order drug release kinetics while the melt-spun rods a zero order one. The bioactivity of the wet-spun rods was tested on lipopolysaccharide-activated BV2 as well as primary murine microglia cells, thus demonstrating the ability of the released flavonoid to favor cell maturation from a pro-inflammatory to an anti-inflammatory phenotype.