Evolution of drug-eluting biomedical implants for sustained drug delivery.

There are a multitude of potential applications for IDDSs, and as a result, there have been many approaches to generate various types of IDDS formulations. IDDSs can be broadly categorized as either passive or active. In the case of passive systems, drug release cannot be controlled once implanted while in the case of active systems, release can be externally triggered [ 3 ]. In this section several types of IDDSs are introduced with examples of their use in both academic research as well as the clinic. For this review, we have intentionally excluded electronic IDDSs such as pumps and microelectrochemical systems, to focus the scope of the article. Readers are encouraged to read previously published reviews on the subject to supplement their knowledge [ 21 , 26 – 30 ]. Figure 1 shows representative images of four of the delivery systems that are discussed. Since the development of the first zero order IDDS in the 1960’s, polymer science has led to the synthesis of more advanced polymeric vehicles for tunable drug release [ 3 ]. Polymers can be distinguished as being either degradable or non-degradable [ 26 ]. Degradable polymer based IDDSs are safer alternatives because they degrade into what are often non-toxic monomers, the metabolized by-products of which are eliminated from the body; and they remove the need for a second surgery to retrieve the implant once the therapy has ended [ 26 ]. The most commonly used degradable polymers are the polyester-based polymers such as poly(ɛ-caprolactone) (PCL), poly(lactic-co-glycolic acid) (PLGA), and poly-L-lactic acid (PLLA) [ 1 , 26 ]. Non-degradable polymers are often biocompatible, inert and drug release is controlled through more simple methods like swelling of the polymer or diffusion; examples of which include urethanes, cellulose derivatives, acrylates, and silicones [ 1 , 26 ]. These materials are ideal for long term orthopedic and dental implants because they are durable with sufficient mechanical strength and biocompatibility [ 3 ]. Polymer based drug delivery systems can be broadly categorized into three general types: monolithic [ 31 – 33 ], reservoir [ 34 – 36 ], or swelling-controlled [ 37 – 40 ]. Monolithic systems are ones where the drug is distributed throughout the matrix. In contrast, reservoir systems have a non-degradable coating that surrounds a drug filled center. Swelling-controlled systems are fabricated from water soluble, cross-linked polymers [ 39 ]. Monolithic systems can be hydrophilic or hydrophobic depending on the composition of the material used. Preparation of a monolithic type system is typically done in one of three main ways: physical mixture of drug and polymer, compression of polymer and drug, or physicochemical activation which can be done without the use of solvents [ 41 ]. One of the main advantages of using a monolithic IDDS is that it is possible to avoid having an excessive burst release of the drug compared to traditional routes of administration and, they are typically easier to produce compared to reservoir type [ 42 ]. The treatment of glioma tumors in the brain using Gliadel® wafers is an example of an effective monolithic IDDS [ 43 , 44 ]. Gliadel® is made from the degradable polymer poly[bis(p-carboxyphenoxy propane) sebacic acid] and loaded with the anticancer drug carmustine [ 44 ]. A meta-analysis looking at newly diagnosed patients versus patients with recurrent high-grade glioma was performed by Chowdhary et al. They found that newly diagnosed patients given the wafers, with or without another adjuvant treatment, had median survival times of 16 months with overall survival rates of 67% and 26% after one and two years, respectively. Patients with recurrent high grade glioma were found to have one and two year overall survival rates of 37% and 15%, respectively with median survival of 10 months [ 43 ]. Holländer et al. used the nondegradable polymer polydimethylsiloxane (PDMS) in their studies combined with the model steroid drug prednisolone. They found that the formulation containing 0.5% prednisolone had the highest cumulative percentage of drug released (16.4–19.9%) after 28 days compared to 9.5–11.0% and 10.4–11.0%. for formulations containing 1 and 1.5%, respectively. They were able to further control the drug release by varying the surface area to volume ratio of the devices. When they increased the pore size, they found that the cumulative amount of drug released increased apart from the 0.5% formulation, which exhibited the opposite trend. They believed that the inconsistency stemmed from the manufacturing process; the drug and polymer were mixed by hand and a degassing step was not performed to remove air bubbles [ 32 ]. Since reservoir type IDDSs are made up of a drug filled core surrounded by a polymer coating, the release rate of the drug is controlled entirely by the polymeric properties. The release profile can be drastically changed by varying polymer properties such as molecular weight and chemical configuration, varying the drug’s physicochemical properties, or by simply changing the thickness of the coating [ 4 ]. These types of systems are ideal for treatments that require long term dosing at a difficult to reach target site or for depot style systemic delivery [ 25 ]. Jadelle®, the modern version Norplant®, is a levonorgestrel-eluting contraceptive which is indicated for a five year period of effectiveness [ 13 , 45 , 46 ]. In a comparative study looking at the etonogestrel-eluting implant Implanon®, Jadelle®, and the copper Pregna® intrauterine device (IUD), researchers found that 0.4% of the women given the implants became pregnant after the third year versus 5.7% for the IUD group [ 45 ]. Lee and Chan demonstrated an alternative insertion technique for the FDA-approved ocular implant Ozurdex® [ 35 ]. This IDDS consisting of the degradable polymer PLGA, releases the steroid dexamethasone (DEX) to the retina and vitreous humor for the treatment of macular edema, diabetic macular edema, and non-infectious posterior uveitis [ 25 , 35 ]. Due to patient variability, this delivery system requires replacement as frequently as every 2 to 6 months [ 35 ]. The goal of IDDSs is to be able to achieve therapeutically effective drug concentrations to the desired site, and in an effort to accomplish this goal, a significant amount of research has gone towards formulating novel polymeric hydrogels able to do just this [ 47 ]. The first use of hydrogels was reported by Wichterle and Lim in 1960 when they fabricated contact lenses using poly(2-hydroxyethyl methacrylate) [ 48 ]. Hydrogels are formed from the 3D crosslinking of hydrophilic monomers which, once formed, can absorb large quantities of biological fluids or water [ 47 ]. The fact that these polymers have a high water content imparts excellent biocompatibility and the ability to load high molecular weight drugs, making them model biomaterials [ 1 ]. The mechanism of drug release is based on the swelling of polymers which typically contain one or more of a combination of carboxylate, hydroxyl, sulfonate, amine, or ether functional groups [ 47 ]. Drug release from hydrogels are controlled by several mechanisms including but not limited to light [ 49 , 50 ], ultrasound [ 51 , 52 ], temperature [ 53 , 54 ], or pH [ 55 , 56 ]. Almeida et al. used the degradable polymer, ethylene vinyl acetate (EVA), with varying percentages of vinyl acetate (VA) and added increasing molecular weights of polyethylene oxide (PEO) as a swelling agent to change the formulation’s swelling capacity. Metoprolol tartrate, a beta blocker, was used as a model drug, and it was determined that drug release was directly related to various parameters including the polymer content, PEO molecular weight, and porosity of the matrix. Generally, they found a direct correlation between VA content and rate of drug release. In one particular formulation, when the PEO (7M) content was increased from 5 to 15%, they achieved 50% drug release after 10 and 3 hours, respectively [ 40 ]. Gustafson et al. reported the synthesis of a nondegradable copolymer of oligo (poly(ethylene glycol) fumarate) and sodium methacrylate to create a biocompatible, charged hydrogel matrix meant for decreasing the risk of infection in orthopedic surgical procedures. They demonstrated the ability to load the antibiotic drug, vancomycin, at a concentration exceeding 500 μg/mg of hydrogel. After 24 hours in Dulbecco’s phosphate-buffered saline (PBS) at 37°C, the formulation achieved release of less than 80% of the total amount loaded [ 57 ]. Injectable hydrogels, or in situ forming implants, use a hypodermic needle to inject the drug and polymer solution or suspension to the target site which is then able to solidify and release the drug in a controlled manner [ 3 , 47 ]. Solorio et al. performed research into a novel in situ forming implant made from PLGA and N-methyl-2-pyrrolidinone and loaded with fluorescein. When comparing the formulation in vitro versus in vivo , they found that there was a greater burst release when formed in vivo . They hypothesized that this was a result of the increased physical pressure occurring in vivo causing increased fluid release from the formulation, and therefore, a higher burst release. They found that this initial burst release was followed by a restricted amount of drug released likely due to porosity changes in the implant reducing the release rate. They also discovered that the implants had 2.1% higher daily release in vitro and a two-fold increase in first order degradation kinetics [ 58 ]. Pakulska et al. were interested in creating a novel hydrogel formulation that would be used to treat spinal cord injuries and deliver therapeutics with the potential for spinal cord regeneration. Their studies involved a hybrid hydrogel which had both chemical and physical crosslinking using a thiol-modified methylcellulose with a poly(ethylene glycol) (PEG)-bismaleimide crosslinker. They chose methylcellulose because of its thermoresponsive properties which allow for the monomers to crosslink and form gels naturally when its temperature reaches 37°C. To test the release profile for this formulation, they combined it with PLGA nanoparticles loaded with the chemokine stromal cell derived factor 1α (SDF1α). They were able to obtain a sustained zero order release of SDF1α over a 28-day period when stored in artificial cerebrospinal fluid at 37°C. When given as an intrathecal injection to a rat spinal cord, they found that the formulation was safe and elicited no behavioral defect to locomotor activity [ 54 ]. IDDSs can be comprised of a non-degradable implanted material which is coated with the drug [ 59 , 60 ]. Oftentimes, polymers are used as the coating vehicle because they offer several advantages including protecting the drug from enzymatic degradation and providing sustained release [ 19 ]. The mechanism by which the drug is released can be generally classified as being either physical or chemical in nature, where chemically-based drug release can be the result of enzymatic or chemical degradation, while physical drug release is dictated by the polymer degrading, osmotic pressure or ion exchange [ 42 ]. Min et al. used a layer-by-layer thin film deposition coating technique on cylindrical polyether ether ketone (PEEK) implants with the degradable polymers poly(acrylic acid) and poly(β-amino esters) (15 and 20 kDa respectively). The top layers contained the broad-spectrum antibiotic drug gentamicin while the deeper layers contained the osteoinductive growth factor, bone morphogenetic protein-2 (BMP-2). The 0.5 – 2 μm thick multi-layered coating released gentamicin with an initial burst release of 60 μg/cm 2 /day on the first day, followed by sustained release (1.0 μg/cm 2 /day) which kept the drug concentration greater than the minimum inhibitory concentration over 40 days. The release of BMP-2 was slower with a rate of 110 ng/cm 2 /day up to day 7, which then decreased to 13 ng/cm 2 /day until day 40 [ 19 ]. Kim et al. studied the dip-coating of a mixture of aspirin and atorvastatin onto a 3D-printed porous PCL vascular stent. The practical application for this system would be to reduce blood low-density lipoprotein cholesterol and the narrowing of blood vessels, or restenosis. They were able to confirm the coating deposition using Fourier transform infrared technology, as well as surface characterization using X-ray photoelectron spectroscopy [ 59 ]. IDDSs using inorganic nanoporous materials such as titania nanotubes [ 61 – 63 ], porous silica [ 64 , 65 ], nanoporous anodic alumina [ 66 , 67 ], and carbon nanotubes [ 68 ] have great potential in the clinical field. These materials are made using various techniques including electrochemical anodization, chemical vapor deposition, and etching of bulk aluminum, titanium and silicone [ 69 ]. These electrochemically engineered systems are able to be specifically tailored down to the length, shape, and diameter of pores to suit delivery needs [ 66 ]. Moreover, the implant surface can be physically or chemically modified in order to improve a drug’s solubility and stability, control drug release, or avoid systemic administration side effects or implant rejection [ 61 , 65 , 70 ]. Another attribute of this type of IDDS compared to polymer based ones is that they are resistant to erosion and do not degrade and have improved chemical, thermal and mechanical stability [ 67 ]. These systems can be combined with other systems such as hydrogels, polymeric vehicles, and other nanomaterials to create more advanced therapies that allow for additional control over the release rate [ 61 , 71 ]. Aw et al. demonstrated the utility of using titania nanotube arrays as a dual delivery vehicle for both hydrophobic and hydrophilic drugs, sequentially. In order to load the drugs, two types of micelles were used: d-α-tocopherol PEG 1000 to load the hydrophobic anti-inflammatory and anti-fungal drugs indomethacin and itraconazole, while the inverted micelle 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N[methoxy PEG-2000] was used to load the hydrophilic antibiotic gentamicin. The two drug filled micelles were loaded into the titania nanotube arrays such that there was no intermixing between the two layers, with the hydrophobic drugs loaded at the opening of the tubes. They demonstrated a sequential release of the therapeutic compounds when the formulation was immersed in PBS at room temperature with a clear separation of the hydrophobic drugs, which emerged over the first five days, from the gentamicin, which was released over the subsequent five days [ 62 ]. Carbon nanotubes, invented by the Japanese physicist Sumio Iijima, are made from rolled graphene sheets and shaped into hollow cylindrical tubes [ 72 , 73 ]. In recent years, research into the use of carbon nanotubes in drug delivery applications has been gradually increasing because of their unique physicochemical properties (such as optical, thermal, and magnetic properties) and structures [ 72 ]. The tubular shape and overall malleability of these nanotubes aids their cellular membrane interaction and ability to permeate different tissues as a result of the “snaking effect” [ 72 , 73 ]. Though carbon nanotubes in their natural state are typically not biocompatible, because of their hydrophobic and nondegradable characteristics, it is possible to overcome these issues through surface functionalization [ 73 ]. Generally the purpose of carbon nanotube functionalization is to physically or chemically increase their aqueous solubility in order to avoid the reticuloendothelial system and combat toxic side effects [ 74 ]. Kavosi et al. evaluated the toxicity and therapeutic efficacy of single- and multi-walled carbon nanotubes (SWCNTs and MWCNTs respectively) on a mouse breast cancer cell line (MC4L2). They found that the carbon nanotubes were able to significantly inhibit cell viability and possessed half-maximal inhibitory concentration (IC50) values of 50 μg/mL after 24 hours and 400 μg/mL after 48 hours for the SWCNTs and MWCNTs, respectively. Flow cytometry results determined that carbon nanotubes induced a time dependent apoptosis in the MC4L2 cells. Histology results from in vivo studies using healthy wild-type female mice dosed with carbon nanotube concentrations of 0.05, 0.25, 0.5 and 1 mg/kg found that the nanotubes caused no histological abnormalities in any of the major organs except the liver which displayed some cell shrinkage compared to the control. They also tested the efficacy of the carbon nanotubes on tumor-bearing mice two weeks after subcutaneous injection of MC4L2 cells. The mice were treated with the chemotherapeutic drug doxorubicin, SWCNTs, or MWCNTs at doses of 2.5, 0.25 and 0.5 mg/kg, respectively. The results of this animal study showed that the average tumor volume was significantly less for the nanotube treated groups compared to the doxorubicin-treated group, with no significant difference seen when comparing the two nanotube groups to one another [ 68 ].

With technological developments and advancements being made constantly, new and emerging IDDSs are fabricated using techniques that were not possible until recently. One example of the future of drug delivery is the use 3D printing to make novel IDDSs [ 10 , 119 ]. For example, Holländer et al. used the fused deposition modelling 3D printing technique to design a T-shaped prototype IUD incorporated throughout with the nonsteroidal anti-inflammatory drug, indomethacin. The implant was made from PCL filaments created using 3D printing to create the IUD. They found that the 3D printed system had a faster release compared to the filaments alone and the mechanism of release was dependent on drug diffusion rather than erosion of the polymer [ 115 ]. Another technology that may play a large role in the future of cardiovascular stents is the use of shape memory polymers which would allow the stent to expand and contract in a temperature-dependent manner [ 42 , 97 , 174 ]. The advantages of using this shape memory technology is that they are capable of degrading into non-toxic byproducts, cost less to produce, are able to improve the stent’s biocompatibility, and require much less work to expand the stent [ 174 ]. One such example of a shape memory polymer being used for this application was demonstrated by Zheng et al. who used a blend of poly(propylene carbonate) and PCL. They showed that the blend of poly(propylene carbonate) and 25% PCL displayed an optimal shape recovery ratio which was 24.1 and 50% greater than formulations made from pure poly(propylene carbonate) or PCL, respectively. The stent was able to transition in shape from a temporary linear form into a permanent spiral at 37°C which lends credit to it being feasible to be clinically relevant. The beta blocker drug metoprolol tartrate, used to manage cardiac arrhythmia and hypertension, was used as a model drug and loaded at 5 wt% of the polymer blend. When immersed in PBS kept at 37°C, the 25% PCL blend displayed a rapid release of 22% in the first 5 days followed by a nearly linear release with approximately 60% of the drug released after 2 months [ 175 ]. The utility of this formulation is that it can be deployed to a constricted vessel while in its temporary linear shape and be activated into its permanent spiral shape solely by body temperature without the need for any additional device such as a balloon catheter which has been linked to adverse effects [ 97 , 175 ] Another advancement that could be made for stents would be to get away from the use of polymers in DESs which have been linked to inflammatory responses caused by the production of ions either from the metal or the polymer [ 42 ]. For example, Farah et al. used a cobalt and chromium-based stent coated with crystallized rapamycin, an immunosuppressive, and looked at the effect of adding a flexible and water-soluble polysaccharide top coating. They found that the top coating technique resulted in significant enhancement in mechanical, physical, and chemical stability and that even after one year there was only about two percent degradation of the crystallized rapamycin. When biocompatibility studies were performed by placing the implant subcutaneously into rats, they found that after one month there were no safety concerns [ 18 ]. Outside of cardiovascular applications, another promising field where IDDSs have utility is for the treatment of infectious diseases such as human immunodeficiency virus (HIV), hepatitis, and tuberculosis (TB) [ 176 – 183 ]. The only commercially available strategy for HIV prevention utilizes antiretrovirals such as the pre-exposure prophylactic, Truvada®. This medication is a once daily oral tablet consisting of 200 mg of emtricitabine and 300 mg tenofovir disoproxil fumarate. The issue with this treatment is that a missed dose dramatically increases infection risk [ 180 ]. To resolve these issues, Johnson et al. fabricated a hollow PCL pellet membrane filled with tenofovir alafenamide and sealed with PCL caps. When the formulation was stored in PBS at 37°C, the drug release rate was inversely proportional to the thickness of the PCL membrane with the 45 and 300 μm membranes having release rates of 0.91 and 0.15 mg/day, respectively. They were able to achieve a sustained release of the drug with 0.28 ± 0.06 mg/day being released over 180 days [ 177 ]. Another IDDS meant for HIV treatment and prevention was created by Barrett et al. to deliver varying concentrations of MK-8591, a nucleoside reverse transcriptase translocation inhibitor with subnanomolar antiviral activity and a long half-life. To formulate the subcutaneous implant, the compound was loaded into polymeric matrices made from either the degradable polymers PLLA or PCL or the nondegradable polymer EVA using hot melt extrusion. In vivo studies were performed first in rats and later non-human primate models and the tested formulations displayed a sustained release which reached clinically relevant drug concentrations for over six months [ 178 ]. The World Health Organization reported that approximately 10 million people contracted TB in 2019 globally [ 184 ]. The work focused on the treatment of osteoarticular TB, the most common type of extra-pulmonary TB, which affects weight bearing bones and vertebrae. In the more severe cases, patients require surgical intervention but even with this technique it is not possible to completely remove the Mycobacterium TB which opens patients to the risk of recurrence and often leads to drug resistance. Since current medications are orally delivered, they must overcome several biological barriers to not only reach the target bone and muscle but also achieve clinically relevant concentrations [ 182 ]. For these reasons, Zhou et al. fabricated hydroxyapatite scaffolds coated with clinical anti-TB drugs, isoniazid and rifampicin. In vivo experiments carried out in an osteoarticular TB rabbit model showed that these scaffolds had greater than 10 and 100-fold higher concentrations in the bone and muscles, respectively, compared to orally dosed animals. They also found that the rabbits had significantly decreased blood drug concentrations and their bones displayed improved regeneration indicating this treatment would decrease the likelihood of systemic adverse effects and impede inflammation associated with TB [ 182 ]. Another avenue that must be discussed when talking about the future of IDDSs is the need for responsive systems capable of dynamic change in drug release rates in response to real time measurements of systemic or local conditions. Further research must be done in order to synthesize biomaterials encompassing specific micro- and macroscopic chemical and structural features which will enable them to be used for this purpose [ 1 ]. The production of novel IDDSs is an interdisciplinary field and as technological advancements continue to occur, the collaboration between biomedical science and innovative computer technology are necessary to create novel systems that were not possible in the past.

IDDSs have utility in a multitude of applications and some systems have reached commercial availability ( Table 2 ). This section discusses several biomedical applications for IDDSs, the main advantages and disadvantages of each are summarized in Table 3 . Heart disease is a serious condition where the patient develops issues with the muscles, coronary arteries or valves of the heart leading to poor coronary function [ 42 , 75 ]. Any disruption to the normal blood flow from the heart means that insufficient oxygen is getting to the body [ 42 ]. In the U. S., it is estimated that 92 million adults have a cardiovascular disease [ 75 , 76 ]. Coronary artery disease is one of the more common cardiovascular ailments and is characterized by the buildup of plaque along the inner surface of the arteries [ 42 , 77 ]. This plaque buildup can cause the arteries to narrow and in extreme cases will lead to a complete blockage [ 42 ]. The first treatment used for coronary artery disease was percutaneous transluminal coronary angioplasty. It involved the use of a catheter with an attached folded balloon which is inflated once the balloon reaches the narrowed artery to compress the blockage against the vascular wall [ 42 , 77 , 78 ]. Drug-eluting balloons offer the ability to deliver the drug locally with a single prolonged inflation and, after providing therapeutic benefit, are removed from the body [ 79 ]. The invention of percutaneous transluminal angioplasty by Thomas Fogarty was meant to help remove blockages in arteries and veins. Further developments made to this technology enabled it to be used to dilate peripheral arteries [ 80 ]. The first clinical application of the drug coated balloon in humans was tested by Scheller et al. who studied the efficacy of the PACCOCATH® balloon catheter coated with the antiproliferative drug paclitaxel (PTX) and contrast media iopromide (Ultravist™) [ 81 , 82 ]. The inclusion of the contrast agent was the result of research performed by Drs. Scheller and Speck who found that this addition increased the drug’s uptake by vascular tissues [ 82 ]. They found that only 5% of patients treated with the drug-coated balloon had restenosis compared to the 43% in the control group [ 81 ]. Unfortunately, using percutaneous transluminal angioplasty increases the risk of thrombosis, or blockage, as a result of the injury made by the catheter and the expansion of the balloon [ 42 ]. The invention in 1977 of the cardiovascular stent reformed the ability to treat coronary diseases [ 42 , 83 , 84 ]. The first generation of stents were mainly fabricated from metals such as stainless steel, platinum chromium alloys, and cobalt chromium alloys and are aptly called bare metal stents [ 77 , 85 ]. Early reports into the safety of bare metal stents showed that 20–30% of patients required intervention within the first 6–12 months due to in-stent restenosis (ISR) [ 42 , 77 ]. The invention of drug-eluting stents (DESs) drastically decreased the rate of ISR to 3–20% [ 77 ]. The danger of using bare metal stents is that there are three stages that occur over the 3 months post implantation where several complications may arise. Between the periods of the first few minutes to a few hours and within a few days to a month, acute or subacute stent thrombosis can occur, respectively. The second stage occurs in one to three months post implantation, when the stent is being surrounded by the vascular tissue but poor or delayed endothelial tissue formation can lead to ISR and thrombosis. The third and final stage can take place three months after implantation where the vascular tissue has fully engulfed the stent, but there is still increased risk of late thrombosis and restenosis due to foreign body syndrome [ 77 , 86 ]. Since bare metal stents require systemic administration of any additional therapeutics, second generation stents which elute drugs are a valuable alternative to combat these risks [ 77 ]. The aim of polymeric coatings for DESs is to increase the ceiling of the amount of drug able to be deposited onto the stent. When polymers are used to coat the stent surface, they provide several advantages. A polymer provides an additional barrier to enzymatic degradation of the drug, keeps the drug attached to the stent surface, regulates the release rate, and, even after the drug is released, the remaining polymer improves the overall biocompatibility of the implant [ 42 ]. There has been some skepticism as to the long term safety of this IDDS because some patients may experience late stage stent thrombosis, and, for this reason, research has been done investigating the inclusion of anti-restenotic drugs to combat this [ 76 , 87 ]. The first commercially available DES, Cypher®, came onto the market in 2002 [ 88 ]. This system was designed to deliver PTX or sirolimus; both of which are antiproliferative drugs however sirolimus is additionally an immunosuppressive compound. The non-degradable polymer coating is comprised of poly (ethylene-co-vinyl acetate) (PEVA) and poly (n-butyl methacrylate) (PBMA). In this formulation, the drug release is controlled by the outermost PEVA/PBMA coating layer which contains no drug and encompasses the inner PEVA/PBMA layer incorporated with the drug [ 42 ]. One of the major drawbacks of nondegradable polymer coated DESs is that it is common for endothelial tissue formation to be incomplete and the host tissue to become overly sensitive to the polymer coating. These side effects meant that delayed healing and late thrombosis were always a potential risk [ 85 , 89 , 90 ]. One avenue that has been extensively studied in order to combat the risk of thrombosis has been the use of the endogenous gas molecule nitric oxide (NO). It has been demonstrated that NO has thromboresistant properties which means the molecule can inhibit platelet activation and adhesion as well as promote wound healing [ 90 ]. Xu et al. demonstrated the utility of NO in their research looking at the use of polyurethane film disks impregnated with 5 and 10% of the NO donor S-nitroso-N-acetylpenicillamine (SNAP). The 5 and 10% SNAP disk formulations were able to slowly release NO at a flux rate of greater than 0.5 × 10 −10 mol cm −2 min −1 for 9 and 19 days, respectively. The SNAP loaded polyurethane disks significantly reduced the platelet adhesion at a rate of approximately 76% compared to unloaded disks. Their results made a strong argument for the utility of NO-releasing polymers as antithrombotic biomaterial coatings for vascular implants [ 89 ]. The issues associated with permanent polymer coatings generated an increased interest in the use of degradable polymers which offer the benefit of a shorter residence time of the polymer in contact with the vascular tissue which has the potential of decreasing the adverse risks associated with permanent polymer DESs [ 85 ]. When non-degradable polymeric materials are present in the body for extended periods, such as the implantation time for a DES, it often leads to adverse effects to the patient, including, but not limited to chronic inflammation, thrombosis, and neointimal hyperplasia, which is the proliferation of vascular smooth muscle cells, causing the thickening of arterial walls [ 42 , 85 ]. The advent of the second generation DESs utilizing degradable polymer-based coatings showed improvements to the pharmacological properties of the DES [ 42 ]. This new formulation had several advantages over first generation DESs including: longer release profiles, versality in types of drugs able to be incorporated, and improved drug loading [ 42 , 85 ]. An example of such a DES is the degradable polymer-coated biolimus-eluting Biomatrix® stent which was used in the LEADERS clinical trial [ 91 , 92 ]. When compared to patients given a first generation permanent polymer sirolimus-eluting DES, it was found that after three years the degradable coating stent had lower incidences of adverse cardiac events (15.7%) (i.e. myocardial infarction or cardiac death) compared to the first generation stent (19%) [ 91 ]. Artzi et al. fabricated a coating for Cinatra™ cobalt chromium stents using cross-linked omega-3 fatty acid, vitamin E, and loaded with the antiproliferative and immunosuppressive drug corolimus. When implanted into the coronary arteries of Yorkshire swine, they found that more than 60% of the total drug amount loaded was released within 5 hours, and by 8 days this value increased to 97% [ 93 ]. With the advent of loading drugs onto the surface of the DESs, came the realization that this type of technology would lead to the initiation of an inflammatory response which would further contribute to ISR [ 83 ]. These stents were considered second generation DESs and still had a permanent backbone [ 94 ]. A newer technology, the third generation of DESs, is still under research and development and utilizes degradable materials to make the stent backbone [ 94 , 95 ]. These DESs are able to provide a temporary support structure to the artery and is able to completely degrade once it has completed its function [ 94 ]. These DESs have been made from various materials including metals such as zinc, iron, magnesium and polymers such as PLLA and PLGA [ 95 , 96 ]. One issue with this type of system is that polymers made from acidic monomers, such as PLGA, will significantly change the pH as it degrades which could damage the surrounding tissue. Furthermore, if the polymer material has inadequate mechanical strength, it may yield to the constricting force of the surrounding vascular tissue and break into small fragments causing downstream vessel blockages, and potentially lead to complications such as stroke [ 97 ]. The first degradable DES implanted in a human was the ABSORB everolimus-eluting stent. It consisted of a PLLA stent coated with poly-D,L-lactic acid (PDLLA) with platinum markers located at either end [ 96 , 98 ]. The rate of major adverse cardiac events was 3.3% after 12 months compared to 14% in a previous study by Tamai et al. using a PLLA degradable stent without drug incorporated [ 96 , 98 , 99 ]. The major adverse cardiac events rate was unchanged even at the five year follow up. They also found that the size of scaffold decreased from 6.94 ± 1.47 to 6.29 ± 1.70 mm after six months confirming the degradation of the stent backbone [ 96 ]. Another example of a degradable DES is the DREAMS 2G stent which is a drug-eluting metal absorbable stent consisting of a magnesium scaffold coated with PLLA incorporated with sirolimus. The BIOSOLVE II international, multi-center clinical trial tested the safety of this DES. It was found that when comparing the quantitative coronary angiography parameters at six and twelve months, the values remained stable and there was no reported stent thrombosis after twelve months [ 100 ]. As stated previously, current treatment options for cardiovascular disease, such as DESs, have the potential to cause adverse effects [ 101 ]. One well accepted treatment strategy proven to be effective for coronary vascular disease, when the affected area is a small diameter artery situated on the heart muscle surface, is the use of coronary artery bypass grafts [ 102 ]. A vascular graft can only be successful when it suspends negative inflammatory reactions and thrombogenesis while simultaneously promoting the necessary endothelialization to avoid these risks [ 101 , 103 ]. Gold standard vascular grafts involve autologous blood vessels, such as the internal mammary arteries and saphenous vein grafts, however both carry disadvantages. The risks associated with these procedures include donor-site morbidity and restenosis. Though several antiproliferative and antimigratory therapeutic agents have been systemically administered with the aim of decreasing the prevalence of these negative side effects, they have had limited success and even shown increased morbidity in extreme cases. For these reasons, it was necessary to create a local, sustained delivery method to better treat coronary vascular disease [ 102 ]. One of the earliest examples of this type of IDDS was created by Edelman et al. in the late 1980s. They synthesized EVA copolymer matrices incorporated with a chemically modified heparin which they implanted adjacent to the carotid arteries of rats. These heparin-loaded matrices demonstrated the ability to effectively inhibit occlusion of the artery while avoiding systemic side effects [ 104 ]. Since EVA is a non-degradable polymer, this type of system was not entirely clinically relevant since it would eventually require a removal surgery [ 102 ]. Since then, several therapeutic agents, which have been used to combat the adverse effects associated with vascular grafts, have been used, and include antiproliferative drugs paclitaxel, sirolimus and sunitinib, antithrombotic agents aspirin and heparin, and antibiotics sisomicin, rapamycin and vancomycin. Examples of polymers that have been used as vehicles for these drugs include PLGA, PLLA, and PCL [ 102 , 105 – 108 ]. More recently, Rychter et al. synthesized electrospun tubules made from the polymers PCL and Pluronic P123 and incorporating varying weight percentages of the platelet aggregation inhibitor cilostazol. When they performed an in vitro release study in PBS at 37°C, they found that the time it took for 50% of the total amount of loaded drug to be released increased as the drug percentage increased from 6.25–18.75% (4.5 and 15 hours respectively) [ 109 ]. Reijnen et al. studied the efficacy and safety of the heparin bonded GORE® VIABAHN® Endoprosthesis when compared to femoropopliteal venous bypass surgery [ 110 , 111 ]. They found that patients implanted with the heparin bonded graft displayed faster recovery rates, improved quality of life, and less morbidity with equivalent patency rates to surgical bypass [ 111 ]. The use of hormonal contraceptives to hinder or postpone pregnancy has been common practice since the advent of the first hormonal therapy in the early 1960’s. The two most widely used methods are: 1) the combined oral contraceptive pill whose active pharmaceutical ingredients are estrogen and progestogen, and 2) the progestogen-only pill whose active pharmaceutical ingredient is progestogen [ 112 ]. Although the pill has displayed >99% effectiveness when used as instructed, it necessitates daily ingestion at precise intervals and, when the dosing schedule is interrupted, the efficacy falls to 92%, and 90–97% for the combined oral contraceptive and progestogen only pill, respectively [ 113 ]. Alternative contraceptive systems have been employed in order to maximize effective hormonal therapy while minimizing the amount of effort on the part of the patient [ 33 , 45 , 113 – 115 ]. Contraceptives that are able to deliver drugs over extended periods of time give patients the ability to receive their treatment via a minimally invasive procedure and reduce the risk of mistakes from incorrect dosing or avoid the expense and regular medical attention required for IUD systems [ 113 ]. The first FDA-approved contraceptive implant, Norplant® in 1990, elutes the progestin hormone levonorgestrel which prevents ovulation by disrupting the uterine lining [ 13 , 113 ]. IDDSs such as Norplant® have a licensed duration of use ranging from 3 to 5 years. It is uncommon for these types of IDDSs to fail, however there are risks associated with them such as ectopic pregnancy, infection at the implant site, and irregular menstrual bleeding [ 116 ]. Since 1990, there has been much research into revolutionary implantable contraceptives. For example, Manoukian et al., using PCL microspheres loaded with levonorgestrel, shaped them into a cylindrical shape using methylene chloride and a stainless-steel mold. The cylindrical microsphere mass was coated with a PCL-based elastomer shell in order to control the release rate of the drug. Although the group did not test the completed implant, they showed that the microspheres alone were able to achieve a release profile that began with a burst release of the drug in the first seven days regardless of drug content (12.5 wt% versus 24 wt%). The longest release profile was found with the 24 wt% formulation which lasted for 56 days when kept at 37°C in PBS compared to 49 and 42 days when the temperature was increased to 42 and 47°C, respectively. This novel IDDS measured 0.2 cm in diameter and 2.0 cm in length with the intention being that it would be injected intramuscularly using a standard hypodermic needle and syringe. The use of this novel microsphere based formulation offers multiple advantages over traditional contraceptive devices namely ease of manufacturing, lower production costs, and feasibility to be scaled up for commercialization [ 113 ]. Another area to address when discussing gynecological implants would be the development of vaginal rings as an IDDS [ 117 – 121 ]. Vaginal rings are able to deliver therapeutic agents in a local and sustained manner and offer the benefit of improved patient compliance [ 119 ]. Since this IDDS is placed vaginally, it offers several advantages over traditional administration routes including a rich blood supply, the ability to bypass hepatic first pass metabolism, and an increased mucosal permeability compared to other tissue types [ 120 ]. This system has been commercially used in applications ranging from vaginal atrophy treatment, such as Pfizer’s silicone elastomer based estradiol-eluting Estring®, to a contraceptive IDDS such as Merck’s EVA based etonogestrel and ethinyl estradiol releasing NuvaRing® [ 119 , 120 , 122 ]. Intravaginal rings have also been utilized to treat the symptoms of endometriosis, a chronic inflammatory disease experienced by 5–10% of premenopausal women. Armed with the knowledge that the aromatase inhibitor, anastrozole, can significantly stifle endometriosis-related pelvic pain, Rotgeri et al. loaded this drug at 35 wt% into a PDMS core which was cut into one of three lengths (1.5, 6, and 38 mm) to give doses A, B and C, respectively. These cores were then surrounded by a PDMS membrane shell and implanted into healthy female cynomolgus monkeys. They found that the rings displayed mean in vivo release rates of 15, 54, and 277 μg/day/animal for doses A, B, and C, respectively over a 42 day period [ 121 ]. An additional example of a gynecological IDDS are the drug-eluting meshes used for pelvic organ prolapse (POP) repair [ 123 – 125 ]. POP is a condition where one or more aspects of the vagina and uterus descend causing the herniation of nearby organs into the vaginal space. In extreme cases, POP can cause dysfunction of the lower urinary tract or bowel and/or sexual dysfunction [ 123 ]. There are several non-surgical and surgical means for POP repair including but not limited to vaginal pessary, hysteropexy, and hysterectomy, however abdominal sacrocolpopexy using either synthetic meshes or biologic grafts is the gold standard [ 123 , 124 ]. These mesh implants have two main objectives, namely, to buttress the damaged soft tissue and provide support for the prolapsed organs [ 125 ]. Numerous complications have arisen from the use of surgically implanted meshes in up to 15% of women [ 126 ] These complications include erosion of the mesh material, infection, chronic pelvic pain, and inflammatory reactions; all of which have a negative effect on a patient’s quality of life and may require corrective surgery [ 124 , 125 ]. In order to decrease the risks associated with vaginal meshes, much research has been devoted to finding materials better suited for pelvic floor implantation as an alternative to conventionally used polypropylene [ 126 , 127 ]. Though there are currently no FDA approved transvaginal mesh products indicated for POP repair available, there has been much research into the development of novel mesh formulations specifically for this purpose [ 123 , 125 – 128 ]. For example, Mangǵr et al. synthesized electrospun PLLA loaded with the hormone 17-β oestradiol at either 1, 5, or 10 wt%. They showed that these meshes triggered the formation of new extracellular matrix components, specifically collagen I, collagen III and elastin, by human adipose derived mesenchymal cells in vitro . Additionally, they demonstrated that these oestradiol releasing meshes doubled the formation of new blood vessels using an ex ovo chick chorioallantoic membrane assay [ 126 ]. Alternatively, Domíngues-Robles et al. made vaginal meshes using thermoplastic polyurethane loaded with varying concentrations of the antibiotic levofloxacin. They confirmed the uniform distribution of the drug throughout the 3D printed meshes which had increased elasticity compared to conventional polypropylene meshes, a beneficial attribute for a vaginal mesh material. Furthermore, the drug loaded polyurethane meshes had significant bacteriostatic activity against Escherichia coli and Staphylococcus aureus with zone of inhibition diameters ranging from 25.2 ± 0.9 to 28.6 ± 0.8 mm [ 127 ]. The typical purpose of an orthopedic implant is to replace joints, stabilize broken bones or misaligned vertebrae, and provide a secure platform for bone fractures [ 129 , 130 ]. Desired properties of an orthopedic implant include: being non-magnetic, possess appropriate mechanical attributes, and be biocompatible or, at the very least, bioinert [ 24 ]. The most commonly used materials for orthopedic implants are made from metals such as titanium and stainless steel, however, alloys using cobalt, chromium and nickel have utility for this application as well [ 130 ]. The most common type of metal-based IDDSs used for orthopedic applications involve incorporating the drug into a coating (either polymeric or ceramic) which is applied to the metal surface [ 71 , 131 – 133 ]. Alternative ways that drug molecules have been coated onto the implant surface include: embedding nanoparticles onto the metal surface, using self-assembled monolayers, or covalent bonding [ 24 ]. One of the main issues associated with orthopedic surgery is the risk of infection, and although advancements in the field over the last two decades have greatly decreased the incidence, the risk accounts for 15% of all hospital-originating infections [ 133 , 134 ]. Infection can cause serious adverse effects resulting in implant failure, and requires surgery either for revision or removal, extended hospitalization which leads to a greater economic burden on the patient, and in the most severe cases, mortality. Mortality rates among patients suffering from periprosthetic joint infection increases from 3.7% to 25.9% when looking at ninety days versus five years from implantation, respectively, compared to patients undergoing revision surgery for aseptic failure (0.8% to 12.9% respectively) [ 135 ]. Conway et al. demonstrated the effectiveness of antibiotic coatings on orthopedic implants in humans when they tested antibiotic cement-coated rods in patients experiencing infected arthrodesis and infected non-union. Sixty percent of patients were cured and, at the last follow up, they reported a 95% limb salvage rate [ 136 ]. Lumbar spine fusion is a procedure used in patients suffering from fractures, instability, degenerative disc disease, deformity, and spinal stenosis [ 137 ]. This type of surgery is association with approximately 2% infection rate most commonly due to Staphylococcus aureus [ 137 , 138 ]. For this reason, Eltorai et al. created a novel pedicle screw and microchip combination ( Figure 2 ) that could be used to deliver the antibiotic vancomycin [ 137 ]. Drug release from the microchip occurs via a micro-reservoir mechanism where voltage applied to the chip results in the dissolution of the reservoir cover allowing for drug release over the course of months [ 137 ]. Though infection is a large concern when it comes to orthopedic implants, implant failure can also be caused by aseptic prosthetic loosening [ 140 , 141 ]. In nearly 10% of all total hip and knee replacement surgery cases, there is a need for revision surgery within the first 10–20 years due to aseptic prosthetic loosening [ 140 ]. Aseptic loosening can be the result of micromotions of the implant relative to the surrounding bone, the creation of particles as a result of the friction between the implant and bone that can cause inflammation and eventual bone resorption, or poor osseointegration, the process by which the implant forms direct contact with the surrounding bone tissue [ 140 , 141 ]. One strategy to overcome this is the use of calcium phosphate (CaP) coatings which are known for their osteoconductive properties [ 140 ]. Kämmerer et al. performed a pilot study using a rabbit model to evaluate the efficacy of the commercially available CaP coating BONIT® to improve osseointegration when it was deposited onto the surface of titanium implants compared to subtractive treated rough surface implants. After two weeks, they found that the bone to implant contact in the cortical bone was significantly higher for the CaP group compared to the rough surface implants (58 ± 7% versus 40.4 ± 18% respectively) [ 131 ]. Pastorino et al. demonstrated the utility of CaP as a drug delivery system in their work using injectable CaP foams. By combining the foam with increasing amounts of the antibiotic, doxycycline, it was shown that the percentage of drug released was inversely proportional to the initial amount of the drug added ranging from 54.88 ± 5.82% to 19.58 ± 2.59%. This phenomenon was likely due to the fact that, in this formulation, increasing the amount of doxycycline led to a loss in the macroporosity of the foam. The fact that the pore size is decreased meant that the volume and, therefore, capacity is decreasing as the drug concentration increases. The novel formulation had a significant bacteriostatic effect on both Escherichia coli and Staphylococcus aureus compared to the formulation with no drug added [ 142 ]. Bone defects in the oral cavity have been treated using various strategies including the use of biological and synthetic materials, including xenografts, allografts, autografts, and alloplastic substitutes [ 143 ]. Although the autograph is the gold standard treatment, it has the risk of donor site morbidity and there are limitations as to the amount of bone available to be harvested [ 144 ]. For these reasons, clinicians desired an alternative treatment method and research has been focused on the development of alloplastic bone substitutes typically made from bioactive glass, calcium sulfate, and CaP [ 144 , 145 ]. Synthetic grafts are normally based on the use of degradable materials in the form of a scaffold [ 144 ]. One of the immediate problems of this type of strategy is the need to control infection caused by bacterial load which can cause immune responses downstream that may result in the rejection of the implant entirely [ 133 , 144 , 146 ]. In the case of antibacterial drug treatment, it is of the utmost importance that the drug concentration does not remain below the therapeutic window, because when bacteria are exposed for an extended period to sub-optimal concentrations, they can develop resistance. In order to avoid this, oftentimes the release profile will be such that there is an initial burst release to rapidly combat the infection after which a lower concentration of drug is released to maintain the concentration in the inhibitory range [ 24 ]. The goal of delivering antibacterial drugs is to minimize bacterial growth and infection at the implant site, however it is very important that the release rate remains within a specific range to be maximally beneficial. When the drug is dispensed too fast there is a possibility that an infection could develop because of the lack of remaining drug in the scaffold. By contrast, releasing the drug too slowly means that the infection has time to spread which further impedes wound healing [ 144 ]. Periodontal disease is an ailment which affects the ligaments, bone and gingivae that help to attach the tooth to the underlying gum tissue. The attachment and eventual growth of gram-positive and gram-negative bacteria create a biofilm and produce compounds which promote an inflammatory response [ 147 , 148 ]. A promising treatment formulation for periodontal disease is the use of in situ forming IDDSs such as Atridox®, Atrigel®, and Alzamer Depot® [ 147 ]. Do et al. investigated several in situ forming IDDS formulations that were made from PLGA, acetyltributyl citrate, N-methyl pyrrolidone or occasionally hydroxypropyl methylcellulose and loaded with the antibiotic minocycline hydrochloride. They found that these in situ forming implants were able to hinder bacterial growth for all ten bacterial strains tested with inhibition zone diameters ranging from 1.7 to 5.3 cm of the petri dish [ 148 ]. Mandibular defects can be the result of inflammatory disease, tumors, trauma, or congenital malformation, and can have a range negative side effects for patients. Since defects of the mandible are often associated with missing teeth, this means a focus for researchers is to improve the success rate for dental implants aimed at replacing missing teeth. The work by Zhang et al. was performed to reconstruct mandibular defects in a mini-swine model by comparing bone graft substitutes made from either biphasic CaP or hydroxyapatite when screwed into place with cylindrical titanium implants. The percentage of bone area formed around the implant was significantly higher for the biphasic CaP group (38.4 ± 5.69%) compared to those treated with hydroxyapatite (13.1 ± 1.61%). There was also a significant difference in the bone to implant contact percentage, 39.4 ± 3.93% versus 18.3 ± 3.84%, for the biphasic CaP versus hydroxyapatite groups, respectively. It was also demonstrated that the force required to pull out the titanium implant from the bone was significantly higher for the biphasic CaP group versus the hydroxyapatite group; 310.4 ± 19.07 N versus 74.6 ± 9.34 N respectively [ 145 ]. Ocular based therapeutics have to overcome several barriers in order to reach the target tissue [ 149 , 150 ]. Topical drug administration accounts for over 90% of all formulations [ 151 ]. Using eye drops or ointments are not ideal for patient compliance because they require multiple applications to maintain the drug concentrations within the therapeutic window due to the low bioavailability of the drug, and the fact that the drug is rapidly washed out into the nasolacrimal duct [ 149 , 152 ]. For this reason, only 1–5% of the total amount of drug being administered to the eye is ever absorbed into the target site [ 149 , 151 ]. In addition to this, in 18.2 – 80% of patients there is an increased risk of infection in the eye through contamination of the eye drop bottle with microbes from the face [ 153 ]. Between 11.3–60.6% of patients do not administer the correct dosage of the drug into their eye [ 152 , 153 ]. The cause for this great variability was found to be a result of several patient factors, including: whether the patient received care at a private institution versus clinic; advanced age; educational level; or disease states of the patients such as arthritis or advanced glaucoma [ 153 ]. One strategy that has been employed to combat patient compliance and mistakes is the use of implantable contact lenses which are able to deliver drugs locally and intravitreally. Using this type of IDDS allows for the drug to have an increased residence time which enhances drug bioavailability, and, in turn, means the drug efficacy increases and side effects can be minimized. For these reasons, much research has been done into the development of a controlled release IDDS formulation for drug delivery to the eye [ 149 ]. Commercially available intravitreal IDDSs include Verisome™, Retisert™, Iluvein™, and Ozurdex™; all of which are currently on the market [ 35 , 150 , 151 , 154 ]. Maulvi et al. developed a contact lens IDDS incorporating hyaluronic acid for the treatment of dry eye syndrome. In vivo testing of the formulation in a rabbit model found that hyaluronic acid was released for up to 15 days when loaded with 80 μg of drug. When they looked into the efficacy of the treatment using benzalkonium chloride to generate dry eye syndrome in rabbits, they found that healing was quicker compared to the control group using the Oxford grading scheme for dry eye syndrome confirmation [ 155 ]. Ocular implants are not limited to implantable contact lenses. Nagai et al., as a technological advancement, manufactured a refillable IDDS made from stainless steel and a PDMS reservoir filled with an injectable gelatin/chitosan gel (iGel) [ 156 ]. For these studies they used either fluorescein or fluorescein isothiocyanate-dextran as a model compound and mixed with the iGel to have a final concentration of 50 mg/mL. The formulation demonstrated a release of fluorescein or fluorescein isothiocyanate-dextran for more than 24 days and 14 days, respectively. They were also able to demonstrate that the release rate of the IDDS overlapped when comparing the profiles for three subsequent refills made to the device [ 156 ]. This IDDS type has also reached clinical relevance in 2020 with Genentech’s Phase III ARCHWAY clinical trial. They are testing the efficacy of a refillable port delivery system for the monoclonal antibody ranibizumab versus intravitreal injections for the treatment of neovascular or “wet” age-related macular degeneration [ 157 ]. The use of cochlear implants is the most common remedy for patients who are suffering from partial to total hearing loss [ 158 , 159 ]. The first recorded application of a cochlear implant was in 1961 by otologist William House and neurosurgeon John Doyle who invented an electrode that was implanted into two patients’ round windows, the opening in the cochlea whose movement activates auditory receptors [ 23 , 160 ]. This early work led to the development of the 3M/House device in 1972 which was formally FDA-approved in 1984 [ 23 ]. Though the outcomes for cochlear implant patients are normally outstanding, there are two situations in which issues can arise, namely post-surgical inflammatory reactions and damage to the spiral ganglion cells [ 158 ]. This damage to the sensory cells of the inner ear can further impede hearing in the patient. For this reason, Eshraghi et al. studied the effect of a DEX-releasing cochlear implant electrode to provide otoprotection against noise trauma. In a guinea pig model, they discovered that the DEX-releasing implant provided a decrease in the auditory brainstem recording threshold shifts when exposed to noise trauma. The DEX-coated electrode also significantly decreased the transcript levels of proapoptotic and pro-fibrotic genes which further demonstrated its otoprotective efficacy [ 161 ]. After the placement of a cochlear implant, the brain learns to translate the electrical stimulation being provided by the implant to discriminate sounds and soundwaves. This process is dynamic, and, for this reason, the electrode must be tuned to accommodate the specific psychophysical threshold and maximum comfortable levels of stimulation (T and C levels, respectively). With time it is normal for changes to occur not only in the T and C levels but also the hardware itself, and this requires reprogramming of the device. It used to be that the only option for reprogramming the implant was to go directly to the audiologist’s office, however, advancements in technology and telecommunication have made it possible for the patient to work with the audiologist without the need to be physically present [ 23 ]. Spinal cord injuries affect nearly 17,000 citizens in the U. S. every year and can ultimately cause permanent paraplegia or tetraplegia [ 162 , 163 ]. When this type of injury occurs, the vertebrae that encircle the soft spinal cord tissue are dislodged and this results in the cord being constricted, killing neurons and glia [ 162 ]. Though there is no treatment for spinal cord injuries, current clinical strategies are aimed at ameliorating pain and managing spasticity [ 162 , 163 ]. More recent studies are looking at the use of therapeutic agents aimed at decreasing inflammation with the goal being to restore lost function. Three therapeutic agents have been in the past or are currently being studied clinically, those being erythropoietin, methylprednisolone, and minocycline. Biomaterials chosen for treating spinal cord injuries must be biocompatible, degradable, and non-toxic [ 162 ]. The first instance of a hydrogel being used in the central nervous system was in the 1990’s, when Plant et al. implanted poly (2-hydroxyethyl methacrylate) hydrogels loaded with Schwann cells into a rat lesioned optic tract model. They were able to demonstrate the axons were able to penetrate into two thirds of the implanted scaffolds [ 164 ]. T.M. Sundt was the innovator who attempted to apply percutaneous transluminal angioplasty to treat basilar artery stenosis after seeing its utility for treating vascular stenoses. He found that of the four patients who received the treatment two years previously, three were disabled from the progressive degeneration while one remained stable [ 80 , 165 ]. Seven years later when different researchers attempted the same strategy, the outcomes were not as positive as desired, likely due to the perforation of arteries supplying blood to the brainstem [ 80 , 166 ]. That was until Rostomily et al. were able to effectively use the procedure in a patient with high grade stenosis of the petrous intracranial internal carotid artery who remained asymptomatic even two years later [ 167 ]. Another avenue that has been explored for brain drug delivery is the need to deliver chemotherapeutic drugs to treat recurrent glioma tumors. In the work performed by Ramachandran et al., they created electrospun nanofiber wafers made from PLGA, PLLA, and PCL, and loaded with the anticancer drug Temozolomide. Using an orthotopic rat glioma model, implanted electrospun wafers displayed a constant drug release of 116.6 μg/day. The formulation that gave a one month release profile resulted in long term survival (greater than four months) of 85.7% of animals [ 168 ].

Since the 1950’s, there have been three main “generations” of drug delivery systems developed. The progression of controllable drug delivery systems began with the first-generation systems mostly consisting of transdermal and oral sustained, controlled release systems. This led to the development of second generation systems where drug release was stimulus-triggered, achieved zero order release, incorporated nanotechnology and were capable of extended release [ 9 ]. The third generation of drug delivery systems are currently in development [ 4 , 9 ]. Although there is no specific classification, examples of this technology include implants that require minimally invasive procedures for insertion and extraction, smaller sized implants, and systems that are used for long term drug therapy treatments [ 4 , 9 ]. In the sixty years since the advent of the Silastic™ material created by Folkman and Long, innovative technologies have enabled IDDSs to become more sophisticated and efficient [ 3 ]. Oftentimes, the purpose of an IDDS is to achieve one or more of the following: 1) reduce side effects, 2) improve the drug’s bioavailability and efficacy, and 3) improve the drug’s solubility and stability [ 1 , 10 ]. Depending on its purpose and location, an implant will either need to be temporary or permanent [ 15 , 16 ]. For example, fixation screws and plates meant to stabilize bone fractures are often removed from the patient once the bone has healed [ 16 ]. In contrast, implants such as pedicle screws and rods used in spinal fusion surgeries, or protheses used in joint replacements are meant to remain for the lifetime the patient [ 17 ]. The versatility of IDDSs permits a wide variety of materials to be used for their construction such as silicone rubber, polymers and hydrogels. These systems can be used for an extensive range of applications such as hormone therapy, antibiotic treatment, and chemotherapy [ 3 ]. IDDSs are useful because they allow for several advantages ( Table 1 ) over more traditional forms of drug delivery and can provide specialized release kinetics [ 1 , 2 , 10 , 18 – 20 ]. One advantage that IDDSs have is that they are well adapted to treat chronic diseases such as diabetes which require long term treatments or dosing at specific intervals. It is also worth noting that traditional drug delivery approaches such as intramuscular or intravenous injections, can result in decreased patient compliance which is something that IDDSs can avoid [ 21 ]. Since there is a direct correlation between poor patient compliance and medication efficacy, the use of IDDSs allows for improved and overall positive outcomes for patients. Between 30–50% of adults suffering from chronic diseases do not follow their prescribed medication schedule. In the US, lack of patient compliance is a problem associated with $100 billion in health care costs and 125,000 deaths annually [ 22 ]. There is a clear need to find a way to improve patient adherence, and the use of IDDSs is a highly promising strategy [ 4 ]. The fact that the implant is placed at the target site means that the drug can circumvent first pass metabolism typically seen with traditional administration routes thereby resulting in an increased amount of available drug [ 10 ]. IDDSs are able to deliver their payload in a localized, controlled and sustained manner that may be programmed by the health care professional [ 3 , 23 ]. These implants can be used to regulate drug release, and therefore have the added benefit of being able to keep drug concentrations within a therapeutically relevant range [ 10 ]. When the drug itself can cause pain or discomfort to the patient, this incident negatively affects a patient’s quality of life and IDDSs can lead to better outcomes. IDDSs have the added benefit of essentially masking the drug from off-target tissue, and instead acting at the desired site thus decreasing toxicity and increasing treatment efficacy [ 21 ]. IDDSs can also be advantageous for treating less common chronic diseases such as neurological disorders, pain treatment and medication abuse [ 4 ].

In conclusion, IDDSs have advanced significantly from the first FDA-approved contraceptive implant, Norplant®. Since then, IDDSs have been made for a range of applications in many fields, including, but not limited to orthopedic, dental, cardiovascular, ocular as well as for spinal cord injuries. As technology advances, the ability to design and manufacture more complex and precise IDDSs is becoming more obtainable. Future work still needs to be done to address issues that still plague modern IDDSs such as the risk of inflammatory responses to the foreign device. In practical terms, IDDSs must be cost effective to produce in order to be commercially relevant. The advantages these types of delivery systems offer such as improved patient compliance, decreased risk of adverse events, and the ability to control the drug release to fit the need are all reasons why these types of delivery systems are the future of drug delivery. It is apparent many more novel IDDSs capable of delivering drugs, in formerly unfathomable ways, are on the horizon.

Effective drug delivery systems are capable of delivering drugs to the target site and maintaining the drug concentration within a therapeutically relevant range [ 1 ]. The dose given must deliver the drug over a specified period of time required to have the most positive effect with minimal negative side effects to neighboring or distal healthy tissue [ 2 ]. Traditional systemic drug delivery using intermittent oral or intravenous administration results in rapid high blood drug concentrations soon after the dose is administered. One problem with this type of drug delivery is that it can be difficult to maintain the desired concentration within a narrow therapeutic window. If the bloodstream concentration of the administered drug is too high, then unacceptable toxic side effects in patients may arise. Alternatively, if the drug falls below the therapeutic level, it is rendered ineffective [ 3 ]. Another issue is that orally administered drugs are subject to first pass metabolism meaning that the drug concentration is substantially reduced, primarily by the liver, prior to reaching the systemic circulation [ 3 , 4 ]. Thus, in order to achieve therapeutic concentrations, patients often must endure multiple dosings and, in the case of intravenous injections, patients may require hospitalization to achieve sustained delivery. Delivering drugs in a localized, sustained and controlled manner can result in patients experiencing minimal side effects [ 3 ]. The advent of the first subcutaneous drug-eluting implant in the 1930’s by R. Deanesly and A.S. Parkes to maintain therapeutically relevant concentrations of hormones including testosterone propionate and oestrone, was the catalyst for an eruption of research in the implantable drug delivery field [ 5 , 6 ]. In the 1940’s R. B. Greenblatt used a subfascially implanted crystalline testosterone propionate pellet to deliver hormone therapy to treat gynecic disorders [ 7 , 8 ]. In 1952, Smith, Kline & French introduced the first controlled release formulation of dextroamphetamine (Dexedrine) initially used to treat depression which was able to achieve a 12 hour release [ 9 , 10 ]. In 1964, Folkman and Long made advancements in controllable drug delivery when they invented silicone rubber capsules, later named Silastic™, to release tri-iodothyronine and isoproterenol [ 11 , 12 ]. This silicone-based system was subsequently used for the levonorgestrel-containing contraceptive Norplant® which became an FDA-approved silicone-based sustained release implantable drug delivery system (IDDS) in 1990 [ 12 , 13 ]. As of 2020, the global market for IDDSs accounts for $21.8 billion and is expected to increase to $31.6 billion by 2025 [ 14 ]. With improvements to fabrication methods and technologies constantly occurring, IDDS production can be performed on a commercially relevant scale. Moreover, companies can be granted extensions to patent applications when creating new products, therefore there is a financial benefit through merging various therapeutics with implantable devices without the need to create an entirely new device. Due to their cost effectiveness and lower overall treatment costs, companies and small start-ups find producing IDDSs profitable, thus patenting novel drug delivery systems is a worthwhile venture [ 4 ]. The aim of this review is to highlight the recent advancements that have been made in the IDDS field and to summarize the various types of research being performed to make novel IDDSs. Aside from discussing the types of systems that have been developed over time, there will also be a focus on commercially available IDDSs; finally, we will look at the future in terms of potential therapeutic benefits to be derived from novel formulations and strategies.