The uses of implantable medical devices are safer and more common since sterilization methods and techniques were established a century ago; however, device-associated infections (DAIs) are still frequent and becoming a leading complication as the number of medical device implantations keeps increasing. This urges the world to develop instructive prevention and treatment strategies for DAIs, producing a publication boom on the design of antibacterial surfaces for implantable medical devices. To help identify the flaws of our current antibacterial designs and advance their clinical translations, the clinical features of DAIs are recently highlighted by Huiliang Cao et al [1].
1. Site-specific Incidence
Infection is a common and frequent complication associated with all types of biomedical materials, despite the infection rate varying greatly among different intended uses of various implantable devices (Table 1) [2-38]. Orthopedic implants, such as the ankle, hip, knee, elbow, shoulder, and finger joint prosthetics, are made of metals (titanium alloys, stainless steel, cobalt-chromium alloy, etc.) and are expected to serve long periods (>10 years) in patients’ bodies. Infections of these devices are extremely troublesome [39]. Ankle arthroplasty has higher infection rates (2.4–8.9%) than hip (0.4–2.4%) and knee (1–2%) arthroplasty, although they are normally made of the same materials (Table 1). This is remarkably related to wound dehiscence (or prolonged drainage) developed due to the frail soft tissue surrounding ankles and increased chance of delayed wound healing following ankle arthroplasty [3,40]. The infection situation will be even more serious in revision cases. For example, the incidence of infection for primary hip and knee arthroplasty is around 2% (Table 1), yet this will be possibly as high as 12% and 22% for revision hip and knee arthroplasty, respectively [41]. Moreover, the number of infection cases is expected to increase progressively because the number of arthroplasty surgeries is going to grow in the coming years. In Taiwan, China, for instance, a total of 728 hip and knee infection cases were recorded in 2013 and this number was expected to increase markedly to over 3500 by 2035 [42]. Not only these metallic implants are connected to bacterial infection, but also polymer devices are susceptible to this complication (Table 1). Examples include breast implants, vascular graft/endograft, cardiovascular electronic devices, and cochlear implants, which are made of silicone, polytetrafluoroethylene, plastics, or Teflon, and have infection incidence high up to 10.2% [9], 6% [10], 7% [14], and 8% [17], respectively. Additionally, the DAIs may occur due to the device design. As in brain stimulation implants, the battery of the pulse generator should be replaced typically every 2 years, and such multiple replacements increase the risk of DAIs [23]. Furthermore, the incidence of infection is highly determined by the site a device is placed in. As shown in Table 1, the infection rates in urinary catheters (up to 13.7 cases per 1000 catheter-days), cerebrospinal fluid shunts (27%), internal fixation devices (32%), and dental implants (47%) are high. This is because these devices are highly challenged by bacterial adhesion and biofilm formation during their insertion and the subsequent service period. For example, urinary catheters provide routes for the entry of pathogenic bacteria, increasing the risk of acquiring infections [28]. Investigations of the bacterial sources in infected shunts also demonstrate that a majority of harmful microbes gained entry from the skin of the patients themselves [43]. The risk of complications in fixation of fractures is highly in connection to the low blood supply and elder people are susceptible to infection [36]. Additionally, there are more than 500 bacterial species associated with commensals or pathogens within the oral cavity [44]. This situation makes the prevention of infections in dental implants extremely complicated. The reported incidence rates for dental implants serving of over 3 and 5 years are 9.25% and 9.6%, respectively, and this rate for implants with service periods of over 10 years is up to 26% [38]. More importantly, the prevalence of the pathogenic strains is also associated with specific anatomical locations. Although Staphylococcus spp. is the most prevalent microbe associated with all types of bacterial infections, other pathogens can be involved in specific sites. Gram-negative microbes are involved in 10-40%, 20%, and 35-55% of vertebral, trauma/fracture, and foot/ankle-related infections [45]. Additionally, 15-30%, 20-30%, and 30-80% of polymicrobial infections occur in vertebral, trauma/fracture, and foot/ankle, respectively [45]. Different bacterial strains may have different metabolisms and pathogenic mechanisms that require specifically tailored treatments. This is especially critical to cure infections involving multiple pathogenic strains; as a result, developing an all-around antibacterial solution for all medical devices is hardly possible.
Table 1. Incidence of typical device-associated infections.
Device |
Materials |
Incidence |
Reference |
Ankle arthroplasty |
Metals (titanium alloys), Ceramic, Polyethylene |
2.4–8.9% |
[2,3] |
Hip arthroplasty |
Metals (titanium alloys, stainless steel), Ceramics (alumina, zirconia), Polymers (polyethylene, polyetheretherketone), Composites |
0.4–2.4% |
[4,5,42] |
Knee arthroplasty |
Metals (titanium alloys, cobalt-chromium alloy), Ceramics (zirconia, titanium nitride), Polymers (polyethylene,) |
1–2% |
[6,42] |
Breast implants |
Silicone |
1–10.2% |
[7–9] |
Vascular graft/endograft |
Polytetrafluoroethylene, Polyethylene Terephthalate, Nitinol |
0.16–6% |
[10] |
Cardiovascular electronic devices |
Plastic polymers, Titanium, Teflon, Gold, Copper |
0.9–7% |
[11–15] |
Cochlear implant |
Teflon, Platinum-iridium alloy, Silicone, Titanium, Ceramics |
1–8% |
[16–20] |
Brain stimulation implant |
Stainless steel, Platinum, Titanium oxide, Iridium oxide |
2–10% |
[21–23] |
Urinary catheters * |
Natural rubber, Polyisoprene, Polymer ethylene vinyl acetate, Polytetrafluoroethylene, Hydrogel |
0.1–13.7 cases per 1000 catheter-days |
[24–29] |
Cerebrospinal fluid shunts |
Silicone rubber |
1.9–27% |
[30–34] |
Internal fixation devices |
Stainless steel, Cobalt-chromium alloys, Titanium alloys |
7–32% |
[35,36] |
Dental implants |
Titanium, Ceramics (zirconia, alumina) |
6–47% |
[37,38] |
* The incidence of catheter-associated urinary tract infection is typically expressed as the number of infections per 1000 urinary catheter-days [29].
2. The Unpredictable Onset
Device-associated infections become even stickier because of those host-specific, transient, or resident factors (Table 2) [46–58]. The onset of DAI is not predictable, it can onset years after implantation (Cases 1 through 6 in Table 2). The soft tissue envelope around an implant likely degenerates with aging and can be disrupted by an occasional scratch, which may have promoted the infection of an alloplastic chin implant 45 years after placement [46]. Breast implants significantly risk bacterial contamination from hematogenous spread of distant antecedent infections. It was reported that the Achromobacter xylosoxidans (lives in wet soil) from a chronic footsore and streptococcus viridans (lives in the oral cavity) from recurrent periodontitis can cause infection of breast implants even 7 and 25 years after the implantation [47]. Staphylococcus epidermidis (S. epidermidis) can colonize various biomedical implants and escape from the immune clearance and antibiotic treatments, hence possibly causing symptom-free (such as pain, redness, or fever) chronic infection lasting even for 30 years before being identified by clinical approaches [46]. Cutibacterium acnes (previously known as propionibacterium acnes), a common conjunctival inhabitant, are slow-growing, anaerobic Gram-positive rods, and can manifest several years or even decades before leading to late infections in orbital implants made of silicone or tantalum [49,50]. The sources of the pathogens of the DAIs can be host-specific (Cases 7 through 9 in Table 2). DAIs can be initiated by acute illness (e.g., diarrhea developed during a holiday journey [8]), penetration of contaminated water during participating in outdoor activities [22], or even when the patients play with their pets (bacterial contamination from zoonotic sources) [51]. Moreover, the occurrence of DAIs is commonly associated with a compromised immune system in the hosts (Cases 10 and 11 in Table 2). Methotrexate, a folate antagonist, can affect neutrophil chemotaxis and induce apoptosis of T cells and reactivation of opportunistic pathogens; hence chronic treatment of rheumatoid arthritis with this kind of drug significantly increases the risk of infections around the battery for brain stimulation [52]. Nocardia nova is a common environmental pathogen and rarely affects immunocompetent hosts; however, this species successfully colonized a tibia implant placed in an immunocompetent patient [53]. Listeria monocytogenes, a common organism associated with unpasteurized dairy products (e.g., deli meats and unpasteurized cheeses), can induce a periprosthetic joint infection in a patient with a history of diabetes mellitus, asthma, and psoriatic arthritis [54]. Anaerobiospirillum succiniciproducens, a common settler in the gastrointestinal tract of cats and dogs, can induce a prosthetic hip joint infection in an immunocompromised patient [55]. DAIs are normally initiated by bacterial seeding and as a result tissue integration will be impaired quickly; however, some cases failed to identify any organism by cultures [56,57] and tissue integration was intact after being infected [58]. These situations add difficulties to the prevention, diagnosis, and treatment of DAIs.
Table 2. Representative cases showing the latent period of DAIs.
Case |
Devices |
Latent Period (Post Insertion) |
Pathogens |
Causes |
Reference |
1 |
Alloplastic chin implant |
45 years |
/ |
After scratching herself (soft tissue degeneration due to aging) |
[46] |
2 |
Breast implant |
Seven years |
Achromobacter xylosoxidans (a pathogen that lives in wet soil) |
Development of a chronic footsore (hematogenous spread from distant bacterial infection sites) |
[47] |
3 |
Breast implant |
25 years |
Streptococcus viridans (a pathogen that lives in the oral cavity) |
After extensive dental treatment (hematogenous spread from distant bacterial infection sites) |
[47] |
4 |
Alloplastic implant |
30 years |
Staphylococcus epidermidis |
Bacterial contamination years before identifying the infection (a symptom-free chronic infection; the pathogen escaped immune clearance and antibiotic treatments) |
[48] |
5 |
Orbital implant |
30 years |
Cutibacterium acnes (previously known as Propionibacterium acnes) |
Bacterial contamination during the primary implantation (the pathogen can manifest for several decades) |
[49] |
6 |
Orbital implant |
26 years (implant exposure 10 years before the presentation was documented) |
Propionibacterium acnes (renamed Cutibacterium acnes) |
Bacterial contamination during the primary implantation or implant exposure during scleral patch graft repair |
[50] |
7 |
Breast Implant |
Five months |
Salmonella serogroup C |
Hematogenous seeding due to developing of diarrhea during a holiday travel |
[8] |
8 |
Generator for brain stimulation |
Four months |
Multispecies including the rare Cupriavidus pauculus species (an environmental Pathogen in “water”) |
Penetration of contaminated water during participating in outdoor activities |
[22] |
9 |
Breast implant |
Seven months |
Pasteurella canis (a pathogen normally lives in the oropharyngeal commensal flora of cats and dogs) |
Bacterial contamination from a patient-owned cat |
[51] |
10 |
Battery for brain stimulation |
Two cases (Two years or 10 years) |
Staphylococcus aureus |
Chronic treatment of rheumatoid arthritis with methotrexate |
[52] |
11 |
Tibia Tenodesis Implant |
Four and half months |
Nocardia nova (a common environmental pathogen, rarely affects immunocompetent hosts) |
Contamination of his tibial wound by the outside facility |
[53] |
12 |
Knee arthroplasty |
4 months |
Listeria monocytogenes (a facultative intracellular organism; commonly associated with deli meats and unpasteurized cheeses) |
Consuming unpasteurized dairy products (an immunocompromised patient) |
[54] |
13 |
Hip arthroplasty |
10 years |
Anaerobiospirillum succiniciproducens (lives in the gastrointestinal tract of cats and dogs) |
Breeding a dog (an immunocompromised patient) |
[55] |
14 |
Knee arthroplasty |
Eight years |
Bartonella henselae (a pathogen that induces acute infections but is hard to be diagnosed by culture) |
A cat scratch |
[56] |
15 |
Cranioplasty implant |
Two years and three months |
No bacteria were cultured, but the infection was clinically evident |
/ |
[57] |
16 |
Shoulder prosthesis |
Three years |
Staphylococcus spp. |
/ |
[58] |
3. Diversity of Relevant Pathogens
Infections associated with medical devices with the same intended use (the same device category) but placed in different individuals are possibly connected with different bacterial strains. As shown in Table 2, the infection of breast implants can result from achromobacter xylosoxidans (Gram-negative rod) [47], streptococcus viridans [47], and salmonella serogroup C [8], or Pasteurella canis [51]. Polymicrobial infections become more prevalent in DAIs [45,59]. Even a single infection in a specific individual often has a polymicrobial composition [60]. Multispecies including the rare Cupriavidus pauculus species were isolated in an infection associated with the generator for brain stimulation [22]. Since the bacteria associated with an infection of a medical device may have diverse morphologies and arrangements, an effective antibacterial strategy must be capable of eliminating multiple pathogenic species. Cocci cells (spherical bacteria) range from 0.5 to 2.0 µm in diameter, rods are approximate 0.5–1.0 µm in width and 1–10 µm in length, and spiral bacteria are up to 20 µm in length and 0.1–1 µm in diameter [61]. Moreover, bacterial morphology varies with the growth environments (medium, surfaces, etc.) and growth phase (normally smallest in the logarithmic phase) [62,63]. These facts add additional difficulties to developing a competent antibacterial surface for implantable devices. On account of these features of DAIs, antibacterial surfaces only have a pore-size-based cell selectivity [64], or those peptide-loaded surfaces merely have specific actions to Gram-positive or Gram-negative strains [65] and are not likely adequate to prevent infection of implantable medical devices.
4. Prevalence of Antibiotic Resistance
The uses of internal implants in humans are safer and more common since sterilization methods and techniques were established at the end of the 19th century [66], and the commercialization of antibiotics especially penicillin in the first half of the 20th century [67]. Antibiotics have become an integral component of contemporary biomedical practice, producing a serious follow-up threat: antibiotic resistance in bacteria [68,69]. Clinical cases in orthopedic practice have shown that infections of antibiotic-resistant bacteria, such as methicillin-resistant Staphylococcus aureus (MRSA), are closely related to high morbidity and mortality [70]. Antibiotic resistance in bacteria even multidrug-resistant (MDR) bacteria is now a worldwide challenge [70]. Antibiotic-resistant infections were frequently reported all over the world, including in both developing and developed countries (Table 3) [71-89]. During an infection, Staphylococcus aureus (S. aureus) often forms biofilms on implantable devices, which dramatically increases the ability of the species to acquire resistance via horizontal plasmid transfer [90]. This is why S. aureus has high rates of resistance. As shown by the typical cases reported in recent years (Table 3), MRSA has become the most common strain causing infections of various implantable medical devices, including cardiac devices [72,74,78,83,85], orthopedic prosthetics [75,76], cochlear implants [77], breast implants [79], laryngeal implants [80], and stent-grafts [81]. In addition, there is an alarming increase in antibiotic resistance in other strains, such as Acinetobacter baumannii [71], Mycobacterium chelonae [73], Enterobacter cloacae complex [81], S. epidermidis [83,89], Klebsiella pneumoniae [84], Staphylococcus haemolyticus [86], and Staphylococcal endophthalmitis [108], are also involved in various resistant DAIs. Those resistant DAIs impacted patients have to experience prolonged hospital stays, bear high medical costs, and risk increased mortality (references in Table 3). Antibiotic recalcitrance is a worldwide threat that likely causes substantial global economic costs ranging from USD 21,832 per individual case to over USD 3 trillion in gross domestic product (GDP) loss by 2050 [91]. In the USA alone, at least 2 million infections and 23,000 deaths per year were caused by antibiotic-resistant bacteria, costing USD 55–70 billion [69]. Currently, antibiotic-loaded materials are important complements to modular medical practices for the prevention of recurrent infections in various medical devices, such as wound dressings, bone cement, bone plates, nails, or prostheses [92-94]. However, applications of these surfaces in “uninfected tissues” to prevent DAIs should be careful and in strict guidance, because the prolonged release of prophylactic antibiotics possibly contributes to arising resistant mutants [95]. Silver-based surfaces also have attractive efficacy in the prevention of DAIs [96], improper use of this material may also pose bacterial-resistant problems [97,98]. In addition, pathogenic bacteria have many defensive actions resistant to antimicrobial challenges [70,99,100]: (a) express polymer biofilms to protect themselves from antibiotic attacks; (b) remodel their outer surface to reduce antibiotic uptake; (c) synthesize precursors to modify the target of antimicrobials; (d) produce enzymes to detoxify dangerous drugs. Therefore, antibacterial surfaces, especially those release-killing ones, should be designed to bypass these actions of bacterial cells.
Table 3. Epidemiology of antibiotic-resistant DAIs.
Case |
Resistant Pathogens |
Implant |
Latent Period |
Reference |
1 |
Multidrug-resistant Acinetobacter baumannii |
Hip arthroplasty |
12–25 days |
[71 |
2 |
Methicillin-resistant Staphylococcus aureus (MRSA) |
Cardiac pacemaker |
Nine years |
[72] |
3 |
Clarithromycin-resistant Mycobacterium chelonae |
Breast implant |
Four days |
[73] |
4 |
MRSA |
Transvenous lead |
Four years |
[74] |
5 |
MRSA |
Ankle fracture fixation |
Eight weeks |
[75] |
6 |
MRSA |
Cranial implant |
Three months |
[76] |
7 |
MRSA |
Cochlear implant |
Five months |
[77] |
8 |
MRSA |
Pacemaker |
Two months |
[78] |
9 |
MRSA |
Breast Implant |
Two days |
[79] |
10 |
MRSA |
Laryngeal implant |
More than one year |
[80] |
11
|
Carbapenem-resistant Acinetobacter baumannii; Fluoroquinolone-resistant Enterobacter cloacae complex (AmpC overexpression) |
Internal fixation for an open proximal tibial fracture |
Two months |
[81] |
12 |
MRSA |
Pacemaker |
Two years |
[82] |
13 |
Multidrug-resistant Staphylococcus epidermidis |
Plates and wire cerclages for periprosthetic fractures |
Three months |
[83] |
14 |
Carbapenem-resistant Klebsiella pneumoniae |
Lumbar instruments, |
Seven days |
[84] |
15 |
MRSA |
The ventricular lead of an implanted defibrillator |
Eight weeks |
[85] |
16 |
Methicillin-resistant Staphylococcus haemolyticus |
Hip joint |
Two years |
[86] |
17 |
Ofloxacin-resistant staphylococcal endophthalmitis |
Intravitreal ozurdex implant |
Three days |
[87] |
18 |
MRSA |
Stent graft |
Three days |
[88] |
19 |
Methicillin-resistant Staphylococcus epidermidis |
Spinal instrumentation |
7–88 days |
[89] |
Because of the aforementioned clinical feature of DAIs, antibacterial designs for implantable medical devices shall conform to a specific intended use. This is crucial to choose proper routes for material synthesis, applicable parameters for material characterizations, and the right strategies for biological evaluations, enabling reproducible, comparable, and reusable results that will be consistent in clinical translation[1,96,101].
References:
Please Note: This entry is a part of a recent paper written by Huiliang Cao et al. J. Funct. Biomater. 2022, 13(3), 86; https://doi.org/10.3390/jfb13030086
This entry is adapted from the peer-reviewed paper 10.3390/jfb13030086