Local Antibiotic Delivery Systems and Their Applications in Orthopaedic Surgery

Abstract

» Local antibiotic delivery systems are critical in managing periprosthetic and fracture-related infections, providing high local antibiotic concentrations without systemic side effects. Polymethyl methacrylate (PMMA), the first material to be mixed with antibiotics to perform local antibiotic therapy in orthopaedic history, offers reliable antibiotic delivery above the Minimum Inhibitory Concentration (MIC), typically used with heat-stable antibiotics such as vancomycin, tobramycin, and gentamicin. However, excessive antibiotics can weaken the cement, especially in total joint arthroplasty (TJA). When used for bone defects, a second surgery is often required to replace PMMA with bone grafts.

» Calcium sulfate (CaSO₄) and hydroxyapatite (HA) combinations provide high antibiotic delivery, with CaSO₄ dissolving over time and CaSO₄/HA offering better bone conversion than CaSO₄. They are most effective in small, contained defects and can be mixed with relatively heat-unstable antibiotics.

» Tricalcium phosphate (Ca₃[PO₄]₂) cements are favored for their biocompatibility and biodegradability, enhancing osteoconductivity and allowing for prolonged antibiotic release. Although clinical studies on Ca₃(PO₄)₂ as an antibiotic carrier are limited, vancomycin is commonly used, showing effective bone formation and infection control with a high bone defect cure and healing rate.

» Hydrogels are 3D networks of hydrophilic polymers that absorb water and can form physical barriers against bacterial agents. Defensive antibacterial coatings (DAC) can be loaded with antibiotics and have shown lower postsurgery infection rates in arthroplasty. DACs reduce bacterial adhesion and can promote bone healing when combined with osteogenic factors, while being bioabsorbable and compatible with living tissue.

» Antibiotic-impregnated bone grafts combine effective local antibiotic delivery with maximum bone healing potential, particularly for those pretreated with induced membranes. Intramedullary harvest offers an unlimited source of bone graft material.

Introduction

Periprosthetic joint infections (PJIs) and fracture-related infections are still major orthopaedic complications despite numerous technical and surgical advances. They account for millions of dollars in costs to the healthcare system^1,2. One of the many advances in recent years to combat orthopaedic infection has been local antibiotic delivery.

Local antibiotic administration roots many years back. When administered locally, higher antibiotic concentrations are achieved without inevitable high-dose parenteral consequences. Although local drug delivery can avoid systemic toxicity, maladjusted dosages can cause local toxicity on skeletal system cells. In orthopaedic settings, polymethyl methacrylate (PMMA) was the first material to be mixed with antibiotics to perform local antibiotic therapy. From there on, biodegradable ceramics such as calcium sulfate (CaSO₄), calcium phosphate, and hydrogels were combined with antibiotics as well^3-5. Numerous studies in the literature have delved into the clinical outcomes of specific local drug delivery systems but only in certain pathologies or procedures. In this article, we attempt to review the current overall usage of PMMA, biodegradable ceramics, hydrogel, and antibiotic-impregnated bone grafts as drug delivery systems in different clinical settings, including their clinical indications, advantages, and disadvantages. The most important clinical studies are summarized in tables for each section, with the largest or highest-level evidence study highlighted in the main text.

PMMA

PMMA has been the most common antibiotic carrier in orthopaedic practice for approximately 50 years^3,6. As beads, spacers, or coated implants, it has significant effects in managing osteomyelitis, nonunion of long bones, and joint arthroplasty surgeries, specifically PJIs^7-12.

Generally, PMMA could provide a favorable local antibiotic concentration above the Minimum Inhibitory Concentration (MIC)¹³. However, the elution time of antibiotics loaded to PMMA cement is extremely variable depending on multiple parameters such as brand, geometry, and porosity of the cement along with some antibiotic characteristics^14-17. Antibiotics impregnated with PMMA should be heat resistant (up to 90°C), chemically stable, water-soluble, and in powder formation (crystalline)^12,18,19. Liquid-formed antibiotics interfere with cement polymerization and weaken the mechanical stability of PMMA spacers^20,21. Moreover, amount and even mixing techniques of antibiotics could alter the elution time in all kinds of PMMA forms and mechanical strength in spacers^13,22. Proper homogenization of the antibiotic and the cement powder is one way to avoid inconsistency of antibiotic elution. Using commercially antibiotic loaded bone cements (ALBCs) is suggested to minimize the interfering effects of manual loading of antibiotics. Yet, each brand and model has its own elution characteristics based on factors such as porosity and viscosity¹². The shape of PMMA also plays an important role. For example, PMMA beads have a higher antibiotic elution compared with structural spacers. This difference roots back to their greater surface to volume ratio. It is suggested to have a proportion of 5% to 10% of antibiotics in PMMA cement to maintain appropriate mechanical stability for temporary spacers¹². Orthopaedic surgeons should be aware that excessive antibiotics could weaken the cement used in TJA^13,18,22-25.

Combination therapy such as vancomycin plus an aminoglycoside such as tobramycin, or gentamicin, is most commonly used in orthopaedic surgeries, which not only covers a broad spectrum of pathogen microorganisms but also induces a synergistic effect by enhancing the elution of antibiotics from PMMA cement via increasing the porosity^12,26-29. Moreover, amphotericin B could be effectively used in suspected or confirmed fungal infections³⁰. However, the elution of amphotericin B is limited; most release occurs within the first 24 hours, with minimal amounts detectable after 1 week³¹. Some studies have listed antibiotics that can be combined with different products of PMMA, sorted by heat stability and elution time, along with recommended dosage, antimicrobial coverage, and important complications^12,13,19.

Ortola et al.³² conducted a retrospective cohort of PJI cases to evaluate the efficiency of ALBC spacers. The authors reported a two-stage surgical method in which they eradicated any systemic or soft tissue infection first and then used prosthesis or arthrodesis alongside antibiotic loaded spacers. Each group consisted of 56 patients and received gentamicin and clindamycin loaded spacers; second group, in addition, received another 4 g of vancomycin. The mean age of the patients was 54.36 and 57.95 years in the first and second groups, respectively. The average duration of follow-up was 32.87 months (at least 8 months for each patient). The rate of reinfection was lower in the second group (1.79% vs 3.7%), but this difference was not statistically significant (p value = 0.53). The number of reimplantations was lower in the first group as well (39 vs 49, p value = 0.04). The need for arthrodesis was higher in the second group (11 vs 4, p value = 0.04). Although novelty of this article is admirable, more studies are needed to prove the better outcomes of the triple antibiotic combination.

Shih et al.³³ conducted a randomized clinical trial to compare the results of systemic antibiotic therapy vs gentamicin PMMA beads in treatment of subacute osteomyelitis. There were 10 and 13 patients in systemic antibiotic therapy and PMMA beads groups, respectively. The mean age was 25 years in systemic antibiotic group and 33 years in PMMA bead group (p value = 0.09). The follow-up period was approximately 55 months. The average time for hospital stay was almost 10 days shorter for PMMA bead group (6.2 vs 16.4 days, p value < 0.001). The total costs were also reduced for PMMA bead patients. One patient had an allergic reaction in systemic antibiotic group. The infection eradication rate was 100% in both groups.

The ability of PMMA in loadbearing is unique among all local drug carriers⁸. Most concerning disadvantages consisted of the requirement for a second surgery to remove the material, local toxic effects on osteoblasts and osteocytes, systemic toxicity (especially regarding aminoglycoside-induced nephrotoxicity), and potentiality of increasing bacterial resistance^34-37 (Table I; Fig. 1).

Fig. 1.

X-rays of the application of PMMA. Fig. 1-A Unstable construct. Fig. 1-B Stable construct.

TABLE I.

Summary of the Major Clinical Studies on the Role of PMMA in Orthopaedic Surgery

First Author	Study Type	Year	Country	Follow-up (Month)	Age (Year)	Sample Size	Material	Procedure	Pathology	Ab Used	Complication	Infection Eradication Rate (%)
PJI
Vasarhelyi et al.38	RR	2022	Canada	174.0	69.0	176	PMMA cement (Simplex)	Stage 1: applying spacer + debridement + irrigation	PJI	Vancomycin 2 g + tobramycin or gentamicin 2.4 g	Recurrence of infection	87.0
								Stage 2: revision arthroplasty
Dias Carvalho et al.39	RR	2021	Portugal	NM	NM	20	Gentamicin loaded PMMA spacer	Stage 1: removal of implant and applying spacer	PJI	Vancomycin 3 g + meropenem 2 g per 40 g	NM	NM
								Stage 2: reimplantation
Ortola et al.32	RC	2017	Italy	32.7	56.1	112	Gentamicin and clindamycin loaded PMMA cement (COPAL G + C)	Stage 1: removal of implant + debridement + irrigation	PJI	Clindamycin 1 g + gentamicin 1 g/clindamycin 1 g + gentamicin 1 g + vancomycin 4 g	Recurrence of infection	96.0/97.0
								Stage 2: debridement + irrigation + implantation of prosthesis or arthrodesis
Radoicic et al.40	RR	2016	Serbia	NM	66.6	18	Gentamicin and clindamycin loaded PMMA cement (Refobacin)	Resection of the implants + debridement + irrigation	PJI	Vancomycin 2 g ertapenem 4 g, and 2–4 g ceftazidime	NM	89.0
Stockley et al.41	RC	2008	UK	74.0	64.0	114	PMMA beads (PALACOS)	Two-stage exchange procedure with antibiotic-loaded cement	PJI	Vancomycin 2 g powder in 40g of cement with gentamycin 1 g in some cases	NM	87.7
Hart et al.42	CS	2006	UK	48.5	68.2	48	CMW cement with gentamicin	Stage 1: Debridement + Abx PMMA	PJI	Vancomycin 1 g for each 40 g mix of cement	NM	88.0
								Stage 2: Removal of spacer
OM
Wu et al.43	RC	2017	China	29.5	41.0	36	PMMA powder/spacer 36	Stage 1: Debridement + Abx PMMA spacer ± flap	OM	Gentamicin 500 mg per 40 g of PMMA powder mixed with 4 g vancomycin powder	NM	83.3
								Stage 2: Removal of spacer, bone graft
Makhdom et al44	RC	2020	USA	40.0	62.0	28	PMMA cement (Simplex)	Limb salvage Nail coated with Simplex cement (Stryker, Kalamazoo, MI)	OM	Vancomycin (2 g in 40 g PMMA powder) + tobramycin (2.4 g in 40 g PMMA powder)	union related or persistent infection	80.0
Wu et al.45	RC	2019	China	29.5	41.0	28	PMMA spacer 20	Stage 1: Debridement + Abx PMMA spacer	OM	NM	NM	100.0
								Stage 2: removal of spacer ± Masquelet bone grafting
Buono et al.46	RC	2018	Switzerland	30.0	41.0	24	PMMA beads 24	Stage 1: Debridement + Abx loaded PMMA beads	OM	Gentamicin	NM	84.6
								Stage 2: Bead removal + bone graft + free muscle
Schade et al.47	RC	2010	USA	NM	61.0	35	PMMA cement (Biomet)	Irrigation and debridement, PMMA ALC insertion	OM	Gentamicin 500 mg; 2.4 g of tobramycin	NM	NM
Shih et al.33	RCT	2005	Taiwan	52.8	30.0	13	Gentamicin-PMMA bead (Septopal)	Open biopsy and surgical debridement + gentamicin loaded PMMA bead implantation	SOM	Gentamicin	None	100.0
Walenkamp et al.48	CS	1998	Netherlands	60.0	NM	100	Gentamicin-PMMA bead	Debridement, intramedullary nailing and gentamicin-PMMA beads	OM	Gentamicin	Lower leg amputation, carcinoma development	92.0
Majid et al.49	CS	1985	Kuwait	NM	27.0	50	Gentamicin-PMMA bead	Radical excision of all necrotic and infected tissues; irrigating cavities with saline and hydrogen peroxide; filled with gentamicin-PMMA beads	OM	Gentamicin	Failure of sinus closure	86.0

CS = case series, NM = not mentioned, OM = osteomyelitis, RC = retrospective cohort, RR = retrospective review, and SOM, subacute osteomyelitis.

Calcium Sulfate

Antibiotic-impregnated calcium sulfate (AICS) products have been used for infection treatment and dead space management of many cases of osteomyelitis, fracture-related infection, and PJI (Fig. 2)^8,50,51. It has superior characteristics compared with PMMA, including high local antibiotic concentration delivery, favorable elution time (about 40-42 days based on the product, beam size, and type of the mixed antibiotics), and potential complete absorption after weeks, which eliminates the need for another surgery to remove the products^52-56. Nonetheless, there is very little evidence to suggest that it is an effective option for managing infected large bone defects, making it difficult to recommend confidently.

Fig. 2.

Management of intramedullary osteomyelitis using absorbable antibiotic calcium sulfate. Fig. 2-A Pre-op X-rays. Fig. 2-B Intra-op images.

Recently, McPherson et al.⁵⁷ used femoral stems coated with AICS in revision total hip arthroplasty. Patients were divided into different subgroups based on the organisms affecting them and the stage of their arthroplasty. There were 111 and 104 patients in the study and control groups, respectively. The overall mean age in both groups was 65.8 years (66.1 in study group and 65.5 in control group). The average follow-up time was 56.6 and 117.5 months in study and control groups, respectively. The mean time for occurrence of PJI was reported to be 14.7 and 9.4 months in the study and control groups, respectively. They reported a significant reduction rate of PJI in the second stage reimplantation subgroup (2/42 or 4.8%) compared with a matched control group (7/30 or 23.3%, p value = 0.02)

Mckee et al. compared AICS with antibiotic loaded PMMA in a randomized control trial. They included 14 patients in each group, with a mean age of 44.85 years. The minimum follow-up period was 24 months (mean 38 months). Infection eradication rate was 86% in both groups. The total number of repeated surgical procedures was 15 and 7 in PMMA and AICS groups, respectively (p value = 0.04)⁵⁸.

Although there is controversy on the safety and clinical efficacy of AICS within the literature^8,59,60, most of the studies concluded that the potential benefits of AICS outweigh the risks^61-64. Even without administration of systemic intravenous antibiotics, AICS had 100% success in infection eradication and united bone formation in 13 cases of infected nonunion of long bones⁶⁵. Nevertheless, AICS usage has been associated with a risk of serosal wound leakage, heterotrophic ossification, or hypercalcemia^66-68. According to the literature, aseptic wound drainage was the most frequently reported complication, ranging from 3.8% to 39.4%^68-72. Some of the associated risk factors for increased wound drainage with AICS include diabetes mellitus, smoking, high CaSO4 bead volume, displacement of the product, recent skin flap surgery, and thin soft tissue coverage by CaSO₄^68,73.

Furthermore, the location of CaSO₄ bead placement can be associated with difference rates of wound drainage⁶⁸. According to the literature, appropriate sites for placement of CaSO₄ beads can be deep hip space, inferior to the acetabulum, in hip arthroplasty, and intramedullary cavity or debridement-induced bone void in long bone osteomyelitis^50,69,74-76. It is noteworthy that subcutaneous migration from deep placement of the CaSO4 beads in arthroplasty procedures can be avoided by watertight closure of deeper spaces^68,77. In chronic osteomyelitis and fracture-induced infections, there was no statistical significance between packing CaSO₄ beads into the intramedullary cavity or directly into the bone defect on the matter of wound leakage⁷⁴.

Most of the commonly used local antibiotics, with the exception of gentamicin and beta-lactam antibiotics, have favorable heat stability at body temperature (37°C) or initial heat generated by curing bone cement⁷⁸. Without hydroxyapatite, bone conversion of CaSO₄ may not be favorable, and it is estimated that it may take up to 12 weeks for bone conversion. This time could be shorter based on the formulation of the product^79,80 (Table II).

TABLE II.

Summary of the Major Clinical Studies on the Role of Calcium Sulfate in Orthopaedic Surgery

First Author	Study Type	Year	Country	Follow-up (Month)	Age (Year)	Sample Size	Material	Procedure	Pathology	Ab used with DAC (antibiotic w/v)	Complication	Infection Eradication Rate (%)
PJI
Piovan ewt al.60	RC	2022	Italy	16.0	64.0	17	Calcium sulfate antibiotic beads (Stimulan)	TKA	PJI	1 g of vancomycin + 240 mg of gentamycin	Not mentioned	88.0
Reinisch et al.63	RC	2021	Switzerland	not mentioned	71.3	27	Calcium sulfate (Osteoset)	D	PJI	Vancomycin (mostly 4 or 6 gr), ceftriaxone, (in 2 cases)	Not mentioned	Not mentioned
Masrouha et al.65	RC	2018	Lebanon	24.0	35.0	13	Calcium sulfate pellets (Stimulan)	D + Bone regeneration + External fixation	infected nonunion of long bones	1 g vancomycin + 240 mg gentamicin	Not mentioned	100.0
OM
Palo et al.54	PC	2023	India	12.0	27.1	95	Calcium sulfate antibiotic beads (Stimulan)	debridement + Bone regeneration	OM	Vancomycin + gentamycin + tobramycin in 3 different groups	Not mentioned	97.8
Jagadeesh et al.62	RC	2022	India	32.2	38.4	50	Calcium sulfate pellets (Stimulan)	D + Bone regeneration + External fixation	OM	2 g of vancomycin	None	88.0
Zhao et al.81	RC	2020	China	18.0	48.2	10	Calcium sulfate	D + Bone regeneration	OM	Vancomycin 300 mg/cc	Delayed wound healing and leakage	Not mentioned
Jiang et al.72	RC	2019	China	26.0	41.0	34	Calcium sulfate (Stimulan)	D	OM	500 mg vancomycin, 80 mg gentamicin	None	85.2
McKee et al.58	RCT	2010	Canada	38.0	44.1	30	Calcium sulfate (Osteoset)	D + implant removal	OM	Tobramycin 4%	Not mentioned	86.0
Other
McPherson et al.57	RC	2024	USA	55.6	66.1	111	Calcium sulfate Coated on femoral stem (Stimulan)	Revision re-implantation	failure of primary THA	1 g of vancomycin + 240 mg of gentamycin	Not mentioned	Not mentioned
Hua et al.82	RC	2022	China	36.0	49.6	32	Calcium sulfate	D + antibiotic loaded granules	Post frx infection	Vancomycin 0.5 g and gentamicin 2 ml	Not mentioned	Not mentioned
Patel et al.83	RC	2022	UK	19.7	38.0	13	Calcium sulfate bullets (Stimulan Rapid Cure)	Stimulan Bullet Mat and Introducer	Frx-related infection	1 g of vancomycin + 240 mg of gentamycin	Not mentioned	100.0
Menon et al.77	RC	2018	India	25.7	51.0	39	High purity calcium sulfate (Stimulan)	D + Bone regeneration	25 COM, 8 infected nonunion, 3 PJI,2 soft tissue, 1 AOM	Vancomycin and/or colistin, vancomycin and gentamycin	None	94.9
Qin et al.74	RC	2018	China	33.7	38.0	35	Calcium sulfate pellets (Stimulan)	D + Bone regeneration + External fixation	Frx-related infection	0.5 g vancomycin + 160 K international unit gentamicin	None	89.0

Frx = fracture, COM = osteomyelitis, D = debridement, OM = osteomyelitis, PC = prospective cohort, PJI = periprosthetic joint infection, RC = retrospective cohort, and RCT = randomized controlled trial.

Hydroxyapatite Combinations

Hydroxyapatite (HA), the primary inorganic component of bone, plays a crucial role in bone regeneration by releasing calcium and phosphate ions⁸⁴. Hydroxyapatite combinations are bioactive and biocompatible ceramics that serve as a key mineral component of bone⁸⁵. HA can be produced both synthetically and through extraction from natural sources. Although numerous methods for synthesizing HA with tailored properties have been extensively developed, the optimal method remains elusive⁸⁶. Owing to the limited mechanical strength of HA and its tendency to fracture, this material is typically incorporated with other biomaterials, such as polyurethane and chitosan, to create a more robust and durable composite⁸⁷. One of these composite materials that has gained significant attention, and is widely used for filling voids and gaps in bones in recent years, is a CaSO₄/HA combination. This is a mixture of 60% α-calcium sulfate hemihydrate, a fast-resorbing material, and 40% hydroxyapatite, a highly osteoconductive material. This compound can be combined with antibiotics, particularly gentamicin, vancomycin, and tobramycin⁸⁸.

Studies show the gentamicin elution from the CaSO₄/HA exhibited an initial burst of roughly 20% within the first 24 hours, followed by a sustained release over the subsequent days, maintaining its antibacterial effect even for microorganisms resistant to systemic use of gentamicin⁸⁷. When gentamicin is used at a concentration of 175 mg/10 ml, the elution can sustain a MIC level for up to 28 days⁸⁹, ensuring effective suppression for bacteria associated with fracture-related infections (according to the European Committee on Antimicrobial Susceptibility Testing). In an in vitro study, Yang et al. found that CaSO₄/HA bone substitute containing 9.76% vancomycin (by weight) released vancomycin after 28 days when an initial dose of 2000 μg was added⁹⁰.

CaSO₄/HA compounds have therapeutic applications that extend beyond joint infections and osteomyelitis, encompassing treatment for fractures, bone tumors, and other bone lesions^91,92. Ferguson et al. conducted a retrospective study on 180 patients with osteomyelitis. Each patient underwent a single-stage surgical procedure and received a combination of CaSO₄/HA and gentamicin. The mean bone defect size was 10.9 cm³. Their study demonstrated a 95.6% eradication rate and a mean bone void healing of 73.9%. Twenty-one patients in this study experienced complications, such as fracture and wound leakage⁹³. Another recent study by Karr et al. on 125 patients with fracture-related infections and osteomyelitis of lower extremities, treated with a combination of CaSO₄/HA, vancomycin, and tobramycin, revealed a remarkable success rate of 96.1%. The maximum size of the defects in this study was 2 to 3 cm³⁹⁴.

CaSO₄/HA can be used in small^95,96 and contained defects^97,98 because of its limited structural strength and inability to support load-bearing applications. Risk of drainage, resistance, and high cost and risk of delamination, resulting in a cascade of complications such as graft dissolution, migration, resorption, heterotopic ossification, and hypercalcemia, are among the potential drawbacks associated with the use of CaSO₄/HA^69,99-103 (Table III).

TABLE III.

Summary of the Major Clinical Studies on the Role of Hydroxyapatite Combinations in Orthopaedic Surgery.

First Author	Study Type	Year	Country	Follow-up (Month)	Age (Year)	Sample Size	Material	Procedure	Pathology	Ab Used	Complication	Infection Eradication Rate (Percentage)
PJI
Jiamton et al.104	PC	2023	Thailand	12.0	47.2	58	Antibiotic impregnated microporous nanohydroxyapatite	D + nHAATB beads application + Gel foam application	PJI	0.5 g vancomycin, In 11 cases, amikacin sulfate (AMK)	wound drainage heterotopic ossification	98.1
Wakabayashi et al.105	RC	2023	Japan	98.0	66.3	13	Antibiotic-impregnated calcium hydroxyapatite (Bone Ceram P; Olympus Terumo Biomaterials Corp, Tokyo, Japan)	D + exchange of the liner and head. cup and/or stem revision were performed with one-stage revision	PJI	60 to 80 mg (vancomycin, cefotiam, Amikacin, flomoxef)	none	75.0
Hasegawa106	PC	2021	Japan	73.0	70.0	14	Bone Ceram P (Olympus Terumo Biomaterials Corp, Tokyo, Japan) antibiotic-impregnated HA	D + two-stage knee revision	PJI	Sealed	none	79.0
Stravinskas et al.107	PC	2018	Lithuania	12.0	76.0	20	Synthetic bone graft substitute (CERAMENT|G, BONESUPPORT AB, Lund, Sweden)	Injection of ceramic material in distal neck + insertion of the sliding screw in trochanteric fractures/injection of c3ramic material into unsupported proximal femoral part in uncemented revision	Trochanteric frx, uncemented hip revision	Gentamicin sulphate (175 mg/10 mL) and vancomycin	NA	100.0
Logoluso et al.88	PC	2016	Italy	18.1	67.8	20	Cerament G or Cerament V, a gentamicin or vancomycin-loaded calcium-based resorbable bone substitute (60% calcium sulphate, 40% hydroxyapatite)	D+ implantation of gentamicin-loaded spacer + revision prosthesis	PJI	Cerament G and Cerament V incorporate gentamicin sulphate (175 mg/10 mL) and vancomycin (66 mg/mL): 10 to 20 mL of Cerament G or Cerament V	Heterotopic ossification, Surgical wound dehiscence	95.0
Choe et al.108	PC	2015	Japan	37.0	66.0	27	Bone Ceram P (Olympus Terumo Biomaterials Corp, Tokyo, Japan) Antibiotic-loaded hydroxyapatite block	D + implant removal in PJI or resection of the femoral head in SA	PJI, septic arthritis	100 mg of vancomycin was loaded, and lid was tightly	NM	100.0
OM
Ferguson et al.93	RC	2023	UK	58.8	51.7	180	Cerament G	D + injectable calcium sulphate/nanocrystalline hydroxyapatite ceramic	OM	Gentamycin	Wound leakage, Fracture	95.6
Henry et al.109	RC	2023	UK	55.8	39.5	81	Cerament G	Fixation	Diabetic Foot OM	Gentamycin	NM	96.7
Kavarthapu et al.110	RC	2023	UK	30.0	56.0	54	21 cerament v 35 cerament G	D + Reconstruction + Cerament application	OM	Vancomycin, Gentamycin	Nonunions + Nonhealing ulcer	96.2
Aljawadi et al.111	PC	2020	UK	22.0	41.2	80	Cerament G	D + F	COM	minocycline (MINO), or micafungin (MCFG)	Superficial wound infection, delayed union, nonunion	96.2
Karr et al.94	RC	2018	USA	74.0	57.0	125	Cerament V + tobramycin	Percutaneous antibiotic delivery technique	COM	Vancomycin	NM	96.1
Drampalos et al.112	PC	2018	UK	16.0	68.0	12	Cerament G	D + local delivery of antibiotic in drilled tunnels using the biocomposite (silo method)	OM	Gentamycin	Wound leakage	100.0
Other
Hoveidaei et al.102	CS	2024	USA	11.6	54.8	21	Cerament G	Frx-related infection/revision arthroplasty	OM, PJI	Gentamicin, vancomycin, tobramycin	Reinfection, drainage, acute sepsis	95.2
Aljawadi et al.113	RC	2022	UK	9.73	42.7	51	Cerament G	Fixation	Open Gustilo-Anderson IIIB Fractures	Gentamycin	nonunion	98.0
McNally et al.114	PC	2016	UK	19.5	51.6	100	Cerament G	Excision + filling with CERAMENT	Gustilo-Anderson IIIB open fractures	Vancomycin. gentamycin	Fracture, wound leakage, Persistent nonunion	96.0

CS = case series, COM = chronic osteomyelitis, D = debridement, F = fixation, Frx = fracture, NM = not mentioned, NA = not applicable, OM = osteomyelitis, PC = prospective cohort, PJI = prosthetic joint infection, and RC = retrospective cohort.

Based on the multiple clinical series listed in Table I, CaSO₄/HA + gentamicin or vancomycin is an excellent choice for contained defects and bone voids. Infection eradication and bone remodeling potential are easily achieved with one surgical procedure with minimal risks for wound drainage and hypercalcemia. However, subtherapeutic levels of antibiotics during prolonged elution could promote the emergence of resistant strains.

Tricalcium Phosphate

Calcium phosphate cements have been known for their favorable biocompatibility, biodegradability, injectability, and biological stability¹¹⁵. Tricalcium phosphate (Ca₃[PO₄]₂) cements are crystalline forms of calcium phosphate cements with a calcium to phosphate ratio of 1.5¹¹⁶. Tricalcium phosphate cements have been widely used as bone void fillers due to their similarity to inorganic bone structure, improving their osteoconductive ability^117-120. Moreover, their ability to maintain a steady, prolonged drug release compared with other local antibiotic delivery materials, such as PMMA, makes them good choices as antibiotic carriers for local antibiotic therapy^121,122.

The elution characteristics of tricalcium phosphate cements depend on multiple factors, including the chemistry of the matrix, porosity, additives, drug types, drug concentrations, drug loading methods, and release media, which have been comprehensively discussed in the review study by Fosca et al.¹²² In contrast to PMMA and CaSO₄, the clinical studies investigating tricalcium phosphate cement as an antibiotic carrier are limited. Vancomycin is the most frequently used antibiotic impregnated to tricalcium phosphate cements in clinical studies^{81,82,123-125}.

In a randomized controlled trial conducted by Lang et al.¹²⁵, involving 98 patients (each group with 49 patients) affected by chronic osteomyelitis (primarily affecting the long bones) with a mean age of 33.14 years, the research group received vancomycin-loaded tricalcium phosphate cement after debridement, while the control group was treated with an irrigation and drainage device for antibiotic delivery. Over the course of 12 months, the research group demonstrated significant healing of bone defects, achieving a high cure rate of 93.87% compared with 73.47% in the control group, along with evidence of tricalcium phosphate cement absorption and osteogenesis. There were no cases of recurrence in the study group, whereas there were11 cases of recurrence in the control group. The reported differences were significantly different between two groups (p value < 0.05). In a retrospective study by Zhao et al.⁸¹ on long bone osteomyelitis, a combination of CaSO₄ and tricalcium phosphate as antibiotic carriers was more effective in bone formation induction and infection eradication with faster resorption properties, compared with CaSO₄ alone. Owing to the scarcity of clinical studies, the major complications and clinical disadvantages are unclear. Most of the experiments are in vitro studies, and more investigation is recommended^122,126.

Hydrogel

Hydrogels are 3-dimensional networks of hydrophilic natural or synthetic polymer chains cross-linked physically or chemically, and they can absorb large amounts of water without dissolving. Defensive antibacterial coatings (DAC) are a new form of hydrogels capable of creating physical and chemical barriers in front of bacterial agents. They are used in a wide array of clinical settings from septic hip to spine degenerative disorders and PJI. They are provided as a powder and can be loaded with sterile antibiotic agents with concentrations between 2% and 10%^127,128.

The use of DAC has been documented for arthroplasty and trauma-associated surgeries. In the field of arthroplasty, the largest study included 373 patients in a multicenter prospective randomized study. All the patients underwent primary arthroplasty for hip (n = 294) and knee (n = 79), and a two-stage revision arthroplasty was used for patients complicated with infection. The treatment group received arthroplasty + DAC and the control group received arthroplasty only. Early and delayed wound healing were not statistically different. However, the rate of postsurgery infection was lower in the DAC group (p = 0.00)¹²⁹. Although this study used a 2-stage protocol for patients with infection, DAC can also be successfully used in 1-stage revision¹³⁰. Data from all the studies are presented in Table IV.

TABLE IV.

Summary of the Major Clinical Studies on the Role of Hydrogel in Orthopaedic Surgery

First Author	Study Type	Year	Country	Follow-up (Month)	Age (Year)	Sample Size	Material	Procedure	Pathology	Ab used with DAC (antibiotic w/v)	Complication	Infection Eradication Rate (%)
De Meo et al.131	Retrospective observational	2023	Italy	34.4 ± 9.4	63.0 ± 24.8	27	DAC	Trauma	Trauma	Vancomycin 2.5%	One case of delayed consolidation of open fracture of tibia and one case of loosening of the hip prosthesis implant	97.3
Pellegrini et al.132	PC	2022	Italy	37.2	69.4 ± 8.3	10	DAC	One-stage revision arthroplasty	Sepsis	Vancomycin 5%, and gentamicin 5%	None	100.0
Parbonetti et al.133	PC	2021	Italy	12.0	61.6 ± 10.6	73	DAC	Vertebral surgery	Degenerative spine disorder	Gentamicin 5%	NM	NM
Zoccali et al.134	CC	2021	Italy	24.0	45.6 ± 21.3	43	DAC	NM	Cancer	Gentamicin, vancomycin, vancomycin + tobramycin, vancomycin + gentamicin	NM	NM
Corona et al.135	PC	2021	Spain	20.1	55.0	8	DAC	Bone defect infection	Femoral bone defect infection	Gentamicin and vancomycin	None	NM
De Meo et al.136	CC	2020	Italy	12.4 ± 5.7	74.9 ± 11.5	34	DAC	Revision arthroplasty	PJI	Vancomycin 2.5%, vancomycin 2.5%, and gentamicin 2%	One systemic disease and one deceased	100.0
Franceschini et al.137	PC	2020		24.0 ± 4.0	NM	28	DAC	Two-stage revision arthroplasty	sepsis	Not mentioned	None	92.8
Zagra et al.138	CC	2019	Italy	32.4 ± 7.2	63.9 ± 11.7	54	DAC	Two-stage revision arthroplasty	sepsis	Vancomycin 5%, teicoplanin 2.5% and ceftazidime 2.5%, vancomycin and rifampicin, vancomycin, and meropenem	None	100.0
Capuano et al.130	CC	2018	Italy	29.3 ± 5.0	71.6	44	DAC	One-stage revision arthroplasty	PJI	Vancomycin 5%, vancomycin 5%+meropenem 5%	None	90.9
Malizo et al.139	PC	2017	Italy, Belgium, France	18.1 ± 4.5	60.5	253	DAC	Frx reduction and internal osteosynthesis	Frx and internal osteomyelitis	Gentamicin 4%, vancomycin 2%	None	100.0
Romanò et al.129	PC	2016	Italy, UK, Greece, Belgium	14.5 ± 5.5	71.0 ± 10.6	373	DAC	Hip or knee arthroplasty	NM	Gentamicin 3.2%, vancomycin 5%, vancomycin 2%+meropenem 2%, teicoplanin 5%, ceftazidime 5%, amphotericin B 5%	None	NM

CC = case control, Frx = fracture, NM = not mentioned, and PC = prospective cohort.

A particularly notable trait of DAC is its ability create a physical barrier that reduces bacterial adhesion at the implant surface, which can prevent massive biofilm formation¹⁴⁰. DAC can be loaded with various antibiotics with concentrations ranging from 2% to 10%¹⁴¹. Applying antibiotics through systems such as hydrogels decreases the need for high-dose systemic therapy. As such, the development of resistance among bacteria will be less likely. Hydrogels are bioabsorbable and do not need a second surgery for removal. Furthermore, they are generally biocompatible with living tissue^142,143 and have thermal stability, which allows them to form a gel consistency at body temperature¹⁴⁴. However, hydrogels do not have any inherent osteoconductive and/or antibacterial effects. However, they can be loaded with progenitor cells or osteogenic factors to promote bone healing. Furthermore, controlled drug release is difficult to achieve with hydrogels in comparison with other biomaterials^142,145,146 (Table IV).

AIBG

Antibiotic-impregnated bone grafts (AIBGs) are an alternative to PMMA and ceramic biocomposites for local antibiotic delivery. Lyophilized bone, a natural carrier, facilitates the controlled release of antibiotics directly into the surrounding tissue¹⁴⁷, achieving remarkably elevated levels of antibiotics, exceeding the MIC required to effectively suppress susceptible bacteria¹⁴⁸.

Antibiotic concentrations consistently exceeded MIC for durations ranging from 14 days to 42 days across various studies on AIBG¹⁴⁹. Antibiotic elution from bone graft is 35 times faster than PMMA¹⁵⁰, which makes it less likely to cause antibiotic resistance¹⁵⁰. Administering high doses of some antibiotics, such as rifampin, minocycline, doxycycline, nafcillin, penicillin, ciprofloxacin, colistin methanesulfonate, ciprofloxacin, tobramycin, and vancomycin and gentamicin, could potentially hinder bone integration and healing^5,151. However, even at high doses, vancomycin did not impair bone healing^152,153. Vancomycin was identified as the antibiotic with the lowest level of osteotoxicity¹⁵¹. No nephrotoxicity was reported for AIBGs, even at high doses of antibiotics^150,154-156. Vancomycin-impregnated allografts exhibited a significantly higher reduction in bacterial growth compared with bone without antibiotics, indicating their enhanced antimicrobial effectiveness¹⁵⁷. Antibiotics penetrate cortical bone less effectively than cancellous bone, leading to lower concentrations of antibiotic elution¹⁵⁸. Collagen constitutes over 90% of the organic component of bone¹⁵⁹. Therefore, in addition to the surface and porous structure of bone grafts¹⁶⁰, the binding affinity of antibiotics to collagen and other proteins plays a significant role in determining the impregnation capacity of the grafts¹⁶¹. Selecting antibiotics with high protein-binding affinity is essential for AIBGs due to their collagen-rich matrix. For instance, cefazolin demonstrates strong protein-binding properties, enhancing its retention in such graft¹⁶². By contrast, PMMA, which lacks proteins, relies solely on its physical structure for antibiotic impregnation, making protein binding less significant in its drug delivery mechanism¹⁶³. Fig. 3 shows the images of AIBG (vancomycin-impregnated) application in a 3-cm infected nonunion tibial defect in a 54-year-old man.

Fig. 3.

X-rays and images of the application of Antibiotic-impregnated bone grafts. Fig. 3-A pre-op. Fig. 3B intra-op. Fig. 3-C post-op.

The infection eradication rate was reported to be superior in the AIBG group compared with the bone graft group (95.6% vs. 82%, respectively)¹⁶⁴. While nonresorbable biologics, including AIBG, may carry a higher risk of reinfection¹⁵⁷, patients undergoing procedures that involve structural AIBG have been associated with improved clinical outcomes^165,166. These findings suggest that AIBGs, if indicated, may be a potential substitute for bone cements but may be more susceptible to bacterial colonization and subsequent reinfection of the surgical site¹⁶⁷.

The potential for an unlimited source of graft material is enhanced with intramedullary graft harvest techniques, allowing for the continuous supply of autologous bone without the significant morbidity associated with traditional graft harvesting methods¹⁶⁸. It is possible to use up to one-third allograft in combination with autograft without compromising the healing process⁴³. This hybrid approach allows for the preservation of native healing capacity while benefiting from the structural and immunological properties of the allograft.

AIBGs offer maximum bone healing potential by combining the regenerative properties of the graft material with localized antimicrobial effects. This synergy facilitates faster integration into the surrounding tissue, enhancing the healing process while significantly reducing the risk of infection. Allograft processing eliminates viable cells, distinguishing them from autografts. As a result, concerns about antibiotic cytotoxicity are less relevant¹⁶⁹. However, the lack of standardization in the type of bone graft, antibiotic choice, dosage, and impregnation method across studies hinders our ability to draw definitive conclusions about the efficacy of antibiotic-impregnated bone grafts (Table V).

TABLE V.

Summary of the Major Clinical Studies on the Role of Antibiotic Impregnated Bone Grafts in Orthopaedic Surgery

First Author	Study Type	Year	Country	Follow-up (Month)	Age (Year)	Sample Size	Material	Procedure	Pathology	Ab Used	Complication	Infection Eradication Rate (%)
Dersch et al.170	RC	2022	Austria	67.2	68.2	70	92 cm3 (range: 40-202 cm3) antibiotic impregnated allograft per surgery, using the “impaction grafting technique”	D + filling the defects	PJI	1 g vancomycin or 0.4 g tobramycin, respectively, in 10 cm3 of bone	NM	92.9
Ebied et al.171	PC	2016	Egypt	72.0	61.0	33	Bone chips driven from grafts	Osteotomy + filling the defects	PJI	Vancomycin 4 mg/head, meropenem 4 mg/head, teicoplanin 400 mg × 4/head, meropenem 2 + vancomycin 2/head	Dislocation, hematoma formation, intraoperative fractures	97.0
Winkler et al.156	PC	2006	Germany	38.0	NM	48	Antibiotic impregnated bone graft	Lying implants were removed + filling the defects	Deep infection of the lower limb	Vancomycin was used in all cases, combinations with tobramycin in cases of mixed infections.	NM	96.0
Chan et al.164	PC	1998	China	58.0	36	46	Antibiotic impregnated bone graft	Debridement + filled with antibiotic-impregnated bone graft	infected tibial nonunions	Vancomycin/piperacillin/Cephalothin/ticarcillin	Osteomyelitis, skin rash, pin tract infection	96.0
Buttaro et al.150	RC	2004	Argentina	32.4	59	29	Vancomycin-supplemented impacted cancellous allograft	Revision THA	Infected THA	500 mg dry powdered vancomycin	Periprosthetic fracture	96.7
Wu172	RC	2011	Taiwan	38.4	33	22	Cancellous bone graft + 3 vials of vancomycin + 3 vials of gentamicin solution	Plate removal + irrigation + filled with Cancellous bone graft	Infected nonunions of the distal tibia	Vancomycin + gentamicin	None	100

RC = retrospective cohort, PC = prospective cohort, PJI = prosthetic joint infection, D = debridement, NM = not mentioned

Conclusion

In conclusion, local antibiotic delivery systems, including PMMA, CaSO₄, HA, tricalcium phosphate, and AIBG, play a crucial role in managing infections in orthopaedic surgery. Each system offers distinct advantages, such as PMMA's sustained antibiotic release, CaSO₄'s complete resorption, and HA's osteoconductive properties. Although these systems are effective in specific clinical settings, ongoing research and careful material selection are essential to optimize treatment outcomes. The continuous evolution of these technologies holds promise for improving patient care by minimizing complications and reducing the need for additional surgeries. Recommendations are presented in Table VI.

TABLE VI.

Recommendations

Recommendation	Grade
PMMA—Provides effective antibiotic delivery with elution above the MIC for up to 4 weeks	C
PMMA—Commonly used heat-stable antibiotics include vancomycin, tobramycin, and gentamicin. Amphotericin B is used for fungal infections	B
PMMA—Avoid using excessive antibiotics when cementing in total joint arthroplasty (TJA), as it weakens the cement	A
PMMA—Typically requires replacement with bone grafts, necessitating a second surgery	A
PMMA—For uncontained defects, it is best to stabilize them with a rod	B
Calcium Sulfate—Use it only for small and contained defects	A
Calcium Sulfate—Low bone conversion rate without HA.	C
Calcium Sulfate/Hydroxyapatite—Mostly effective in small, contained defects and can be mixed with relatively heat-unstable antibiotics	B
Calcium Sulfate/Hydroxyapatite—Be careful about wound leakage as it’s one of the most common adverse effects	B
Tricalcium Phosphate Cements—Effective bone formation and infection control with a high bone defect cure and healing rate in osteomyelitis (limited studies)	A
Hydrogel—The most used antibiotic in the literature is vancomycin	B
Hydrogel—DAC can be used for both fracture-related infections and one- or two-stage septic arthroplasty revision surgeries	B
Hydrogel—DAC prevents massive biofilm formation	C
AIBG—It is possible to use up to one-third allograft in combination with autograft without compromising the healing process	B
AIBG—Vancomycin was identified as the antibiotic with the lowest level of osteotoxicity. However, high antibiotic doses, toxic to osteoblasts, could potentially hinder bone healing	C

DAC = defensive antibacterial coatings, PMMA = polymethyl methacrylate.

According to Wright¹⁷³, grade A indicates good evidence (Level I studies with consistent findings) for or against recommending intervention; grade B, fair evidence (Level II or III studies with consistent findings) for or against recommending intervention; grade C, poor-quality evidence (Level IV or V studies with consistent findings) for or against recommending intervention; and grade I, insufficient or conflicting evidence not allowing a recommendation for or against intervention.