Abstract
Background
Low-dose antibiotic-loaded acrylic cement is routinely used for preventing skeletal infection or reimplantation in patients with periprosthetic joint infections. However, few reports about the selection of antibiotics in acrylic cement for antigram-negative bacteria have been proposed.
Questions/purposes
(1) Does the addition of antibiotics (tobramycin, meropenem, piperacillin, ceftazidime, ciprofloxacin, and aztreonam) to acrylic cement adversely affect compressive strength before and after elution? (2) Which antibiotics have the highest cumulative release within 28 days? (3) Which antibiotics showed antimicrobial activity within 28 days? (4) Does meropenem-loaded cement improve body weight, temperature, and other inflammatory markers compared with control unloaded cement?
Methods
This is an in vitro study that assessed the mechanical strength, antibiotic elution, and antibacterial properties of antibiotic-loaded cement, combined with an animal study in a rat model that evaluated key endpoints from the animal study. In the in vitro study, we added 2 g of tobramycin (TOB), meropenem (MEM), piperacillin (PIP), ceftazidime (CAZ), ciprofloxacin (CIP), and aztreonam (ATM) to 40 g of acrylic cement. The compressive strength, elution, and in vitro antibacterial properties of the antibiotic-loaded cement were detected. Thirty male rats were randomly divided into two groups: CON (antibiotic-unloaded cement) and MEM (meropenem-loaded cement, which had the most stable antibacterial properties of the six tested antibiotic-loaded cements in vitro within 28 days). The right tibia of all rats underwent arthroplasty and was implanted with the cement, followed by inoculation with Pseudomonas aeruginosa in the knee. General status, serum biomarkers, radiology, microbiological assay, and histopathological tests were assessed over 14 days postoperatively.
Results
The compressive strength of all tested antibiotic cement combinations exceeded the 70 MPa threshold (the requirement established in ISO 5833). The cumulative release proportions of the raw antibiotic in cement were 1182.8 ± 37.9 µg (TOB), 355.6 ± 16.2 µg (MEM), 721.2 ± 40.3 µg (PIP), 477.4 ± 37.1 µg (CAZ), 146.5 ± 11.3 µg (CIP), and 372.1 ± 14.5 µg (ATM) within 28 days. Over a 28-day period, meropenem cement demonstrated antimicrobial activities against the four tested gram-negative bacteria (Escherichia coli, P. aeruginosa, Klebsiella pneumoniae, and Proteus vulgaris). Ciprofloxacin cement inhibited E. coli growth, ceftazidime and aztreonam cement inhibited K. pneumonia growth, and tobramycin cement inhibited P. aeruginosa. Only meropenem demonstrated antimicrobial activity against all gram-negative bacteria on agar diffusion bioassay. Rats treated with meropenem cement showed improved body weight (control: 280.1 ± 4.2 g, MEM: 288.5 ± 6.6 g, mean difference 8.4 [95% CI 4.3 to 12.6]; p < 0.001), improved knee width (control: 13.5 ± 0.3 mm, MEM: 11.8± 0.4 mm, mean difference 1.7 [95% CI 1.4 to 2.0]; p < 0.001), decreased inflammatory marker (control: 316.7 ± 45.0 mm, MEM: 116.5 ± 21.8 mm, mean difference 200.2 [95% CI 162.3 to 238.2]; p < 0.001), decreased radiographic scores (control: 17.7 ± 2.0 mm, MEM: 10.7± 1.3 mm, mean difference 7.0 [95% CI 5.4 to 8.6]; p < 0.001), improved bone volume/total volume (control: 8.7 ± 3.0 mm, MEM: 28.5 ± 5 .5 mm, mean difference 19.8 [95% CI 13.3 to 26.2]; p < 0.001), decreased Rissing scale scores of the knee gross pathology (control: 3.3 ± 0.5, MEM: 1.1 ± 0.7, mean difference 2.2 [95% CI 1.7 to 2.7]; p < 0.001), decreased Petty scale scores of knee synovium (control: 2.9 ± 0.4 mm, MEM: 0.7 ± 0.7 mm, mean difference 2.1 [95% CI 1.7 to 2.5]; p < 0.001), and decreased bacterial counts of the bone and soft tissues and negative bacterial cultures of cement (p < 0.001, p < 0.001, p < 0.001, p < 0.001, respectively).
Conclusion
In this current study, MEM cement had the most stable in vitro antimicrobial activities, effective in vivo activity while having acceptable mechanical and elution characteristics, and it may be an effective prophylaxis against skeletal infection caused by gram-negative bacteria.
Clinical Relevance
Meropenem-loaded acrylic cement is a potentially effective prevention measure for skeletal infection caused by gram-negative bacteria; however, more related clinical research is needed to further evaluate the safety and efficacy.
Introduction
The incidence rate of gram-negative bacteria in the skeletal infection ranges from 6% to 48% [1, 2, 5, 11, 21, 23, 24], and the common gram-negative microorganisms include Escherichia coli, Pseudomonas aeruginosa, Klebsiella pneumonia, and Proteus vulgaris. Several researchers have evaluated the related properties of low-dose, antibiotic-loaded cement against gram-negative bacteria, such as mechanical, elution, or antibacterial capacity. Gandomkarzadeh et al. [8] studied the mechanical properties of ciprofloxacin cement in the form of 2.5 wt% and 5 wt% of bone cement and suggested that the compressive and bending strength of 5 wt% cement was lower than the minimum ISO values, whereas the 2.5 wt% group met the requirements. Slane et al. [24] reported that the cumulative elution of 3.75 wt% tobramycin cement was 0.33 mg at 28 days, and the compressive strength was above 70 MPa. Samuel et al. [25] studied the elution of 1.25%, 2.5%, and 5% weight of meropenem cement and indicated that the proportion of the cumulative release in the raw antibiotics accounted for 0.58%, 0.83%, and 1.2% for 34 days; the antibacterial activities of the 5% weight samples against P. aeruginosa, E. coli and K. pneumoniae lasted for 3 weeks.
However, the previous studies are almost all in vitro studies of a single antibiotic-loaded acrylic cement, and most studies focused on one aspect, such as mechanical properties, elution, or in vitro antimicrobial activities. No comparative studies between various antibiotic-loaded bone cements have been reported to our knowledge. Up to now, the best possible antibiotic against gram-negative bacteria-loaded bone cement has not been evaluated. Currently, the evidence suggests that meropenem (MEM), ceftazidime (CAZ), piperacillin (PIP), ciprofloxacin (CIP), aztreonam (ATM), and tobramycin (TOB) may be the most common antibiotics targeting against gram-negative bacteria with relatively few reports of drug resistance [6, 7, 27, 28].
Therefore, we asked: (1) Does the same amount of antibiotics added (tobramycin, meropenem, piperacillin, ceftazidime, ciprofloxacin, and aztreonam) to acrylic cement adversely affect compressive strength before and after elution? (2) Which antibiotics have the highest cumulative release within 28 days? (3) Which antibiotics showed antimicrobial activity within 28 days? (4) Does meropenem-loaded cement improve body weight, temperature, and other inflammatory markers compared with control unloaded cement in a rat model?
Materials and Methods
The animal study was designed and performed following the Animal Research: Reporting of In Vivo Experiments and the Institutional Animal Care and Use Committee guidelines. We used 30 male specific pathogen-free grade Wistar rats. The animals, which were 10 weeks old and weighed 250 g ± 6 g, were grouped in cages based on the type of treatment with enriched materials and maintained at 22°C ± 1°C (humidity: 50% ± 10%) over a 12-hour light/dark cycle. Rat welfare was checked daily and the animals had free access to food and water. Clinical orthopaedic acrylic cement (Simplex P, Stryker) was used for all tests. Analytical-grade antibiotics (MEM, CAZ, PIP, CIP, ATM, and TOB) were purchased from Aladdin and used for preparation of the antibiotic-loaded cement specimens. Antibiotic loading is represented as the weight percentage (wt%): the weight of the antibiotic powder as a percentage of the 40-g cement powder. Standard-grade antibiotics were purchased from the National Institutes for Food and Drug Control of China and used in the determination of high-performance liquid chromatography to calculate the concentration and cumulative release of analytical-grade antibiotics in the eluent of cement samples. Figure 1 shows the experimental design of the study.
Antibiotic-loaded Acrylic Cement Specimen Preparation
We added 2 g of antibiotic powder to 40 g of cement powder (5 wt%) with hand-mixing (three stirs per second for 5 minutes) under atmospheric conditions. The amount of antibiotic in the commercial antibiotic-loaded acrylic cement is only 0.5 g to 1 g per 40 g acrylic cement; manual addition of antibiotics would affect the mechanical properties of acrylic cement, depending on the amount and type of antibiotics added. In general, adding 2 g of antibiotics to 40 g of acrylic cement powder usually results in a satisfactory release concentration of antibiotics and acceptable mechanical properties [3].
A liquid monomer was then added into the antibiotic-loaded cement mixing powder and mixed manually (three stirs per second for 60 seconds) at a temperature of 23°C ± 1°C and humidity of 50% ± 10% according to the manufacturer’s instruction. On reaching the dough phase, cement was pressed into stainless steel molds and allowed to cure for 60 minutes. The tested cylindrical specimens had uniform sizes (height: 12.00 ± 0.05 mm, diameter: 6.00 ± 0.05 mm) (Fig. 2A-B). Bone cement cylinders in vitro or MEM-cement samples in vivo were separated and sterilized with ethylene oxide (sterilization temperature: 55°C to 60°C under negative pressure; sterilization time: 1 hour; residual removal time > 12 hours; the entire process lasted longer than 15 hours). The samples were sealed and stored after sterilization and removed with aseptic operation in a laminar flow clean bench before use for the experiment.
Fig. 2.
A-D This figure shows in vitro sample preparation, compressive strength test results, and results of our analysis of antibiotic acrylic cement. (A-B) These photographs show preparation and measurement of compression and elution-tested antibiotic cement samples (height: 12 ± 0.05 mm; diameter: 6 ± 0.05 mm). (C) This photograph shows the in vitro compressive strength test of antibiotic acrylic cement using a material testing machine (Instron 5969) and the compression-tested sample before and after testing. (D) This graph shows results (mean ± standard error of the means) of compression testing nonelution and after elution (n = 8); CON = control. ap < 0.05 (compared with all antibiotic cement groups), bp < 0.05 (compared with other noneluted antibiotic cement groups), cp < 0.05 (compared with other eluted antibiotic cement groups). A color image accompanies the online version of this article.
Fig. 1.
This figure shows the experimental design of the current study.
Mechanical Properties
Specimens (eight per group) were allowed to cure for a further 24 hours before mechanical testing. Another eight specimens per group were soaked in phosphate-buffered saline at 37°C for 4 weeks to elute the antibiotic. These wet specimens were air dried and mechanically tested in the same manner as the 24-hour specimens. Compression tests of the cylindrical specimens (diameter: 6 mm, height: 12 mm) (Fig. 2A-B) were performed using an Instron 5969 material testing machine (Instron Corp), with a crosshead rate of 20 mm/minute (strain rate = 2.0 × 103 s1) between stainless steel plates (Fig. 2C).
Antibiotic Elution
To determine antibiotic release from the cement specimens (six per group), we submerged cylindrical specimens (height: 12.00 ± 0.05 mm, diameter: 6.00 ± 0.05 mm) in 5 mL of phosphate-buffered saline solution and stored them in an incubator at 37°C with constant shaking at 60 rpm [24]. Daily antibiotic release was measured at seven timepoints (days 1, 3, 5, 7, 14, 21, and 28), and the cumulative release over 28 days was calculated. Samples were removed from the fluid, gently washed with phosphate-buffered saline, and transferred to 5 mL of fresh phosphate-buffered saline to maintain sink conditions. Eluent from the previous timepoint was collected and stored in cryotubes at -80°C before analysis. Antibiotic concentrations in the collected eluents were quantified using high-performance liquid chromatography–mass spectrometry (Thermo TSQ Quantis).
Modified Microtube Dilution Bioassay of Elution Samples
A modified microtube dilution bioassay was used to measure the biological activity of the released antibiotics (five per group) in the sample aliquots based on the previous studies [4, 10]. Briefly, the elution samples (10 µL per well) at the special time points (days 1, 3, 5, 7, 14, 21, and 28) were inoculated with tested bacteria (E. coli ATCC 25922, P. aeruginosa ATCC 27853, K. pneumoniae CMCC 46117, and P. vulgaris CMCC 49027) of 106 colony-forming units (CFUs)/mL (100 µL per well) in 96-well cultured dishes and incubated at 37°C for 24 hours. The growth of bacteria was compared visually among the antibiotic-loaded cements and against the positive control (without the antibiotic). The positive control (without the antibiotic) was cloudy due to bacteria growth, whereas the wells containing the eluted antibiotic became clear depending on the concentration and antibacterial abilities of the antibiotic.
Kirby-Bauer Bioassay of Elution Samples
The bioactivity of the released antibiotics (five per group) was determined using an agar disk diffusion bioassay [10]. Briefly, disks (PDM Diagnostic Disks; without any drug; AB Biodisk; diameter: 6 mm, thickness: 1 mm) containing 20 µL of the elution sample at the special time points (days 1, 3, 5, 7, 14, 21, and 28) were placed on a Mueller-Hinton plate swabbed uniformly with 106 CFU/mL tested bacteria (E. coli ATCC 25922, P. aeruginosa ATCC 27853, K. pneumoniae CMCC 46117, and P. vulgaris CMCC 49027). The disks were incubated at 37°C for 24 hours, and the inhibitory activity on the disks (diameter of the inhibition zone) was determined using precision electronic vernier calipers. No antibacterial activity was set at 6 mm (diagnostic disk diameter), inhibition zone diameter of the released antibiotics was set as measured value minus 6 mm, and inhibition zones with a width 3 mm or less were scored as no antimicrobial activity based on the previous study [7].
Surgical Procedure
For the surgery, the rats were anesthetized using 2.5% inhalational isoflurane delivered via a nose cone. After the rats were anesthetized and shaved, we disinfected the skin of the right leg, surgically exposed the knee, and drilled a 2.0-mm hole into the tibia canal. Thirty rats were randomly divided into two groups: CON (antibiotic-free cement, 15 rats) and MEM (meropenem-loaded cement, which was the most stable of antibacterial properties among the six tested antibiotic-loaded cements in the in vitro study, 5 wt%; 15 rats). The cement (diameter: 2.0 mm, length: 8 mm) was manually placed through anterograde insertion (Supplementary Fig. 1A-B; http://links.lww.com/CORR/A922). After closing the joint capsule, we injected 20 µL of 1.5 × 106 CFU/mL suspension of P. aeruginosa (ATCC 27853) into the articular cavity (Supplementary Fig. 1C; http://links.lww.com/CORR/A922) based on the previous animal model of P. aeruginosa infection study [26]. Radiographs were obtained immediately postoperatively to confirm cement position (Supplementary Fig. 1D; http://links.lww.com/CORR/A922). Pain was controlled with buprenorphine within 3 days postoperatively (0.1 mg/kg/day). All rats survived the surgery and the postoperative period without signs of systemic illness. All animals were euthanized 14 days postoperatively for tissue harvest.
General Status and Immune Response
Body weight, body temperature, and the maximal medial-lateral knee width (as described by Miller et al. [20]) of the rats (15 per group) were measured preoperatively (day 0) and on days 1, 4, 7, and 14 postoperatively. Serum was analyzed for circulating levels of serum alpha-1-acid glycoprotein (α1-AGP), an acute-phase reactant in rats and humans that rapidly upregulates within 1 to 2 days in response to infection [5, 16, 18]. The serum levels of α1-AGP (six per group) were measured preoperatively (day 0) and on days 7 and 14 postoperatively using an ELISA (Cusabio).
Radiography
Radiographs (10 per group) were taken on day 14 using the Bruker Xtreme BI (filter: 0.4 mm, 45 kvp, exposure time: 1.2 s, bin: 1×1 pixels, field of view: 10 cm, F Stop: 2). Each radiograph was assessed by two experienced observers (HX, HJS; not study authors) blinded to treatment based on a system used in the previous studies [14, 15, 23]. The scores from the two evaluators were averaged for statistical evaluation.
Micro-CT
Tibial bone (five per group) was scanned and analyzed by Skyscan1276 Micro-CT system using the following settings: voltage: 65 kV, filter: 0.5 mm, current: 200 µA, exposure time: 400 ms, image pixel size: 20.31 µm. The region of interest, namely 0.5 mm below the growth plate on the proximal tibia with a height of 3 mm, was selected for bone microstructure analysis. Herein, bone volume per trabecular volume (BV/TV) was used for evaluating.
Histopathologic Evaluation
Based on the modified Rissing scale score [22], gross pathology (10 rats per group) was evaluated by two experienced observers blinded to treatment (GFW, XPW; not study authors). The gross pathology was determined by grading tissue destruction from 0 to 4. A score of 0 represented the absence of abscess, sequestrum, active bone formation, and erythema. A score of 1 indicated minimal erythema without abscess or evidence of new bone formation; a score of 2 indicated erythema with a widening of the head and shaft of the bone with new bone formation; a score of 3 indicated abscess with new bone formation, sinus tract drainage, or grossly purulent exudate; and a score of 4 typically indicated severe bone resorption, abscess, and diaphyseal or total tibia involvement. Dehydrating wax was applied and the samples were embedded in paraffin, and the samples (synovium, 10 per group) were sectioned (4 µm) and stained with hematoxylin and eosin. All slices were observed and photographed using the H550S Photo Imaging System (Nikon). For histologic scoring of the soft tissue, we referred to the modified Petty scale [12, 13, 29] as follows: score 0 (absent), absence of inflammatory cells; score 1 (mild), presence of occasional polymorph nucleated leukocytes; score 2 (moderate), scattered polymorph nucleated leukocytes and microabscesses; score 3 (severe), diffuse polymorph nucleated leukocytes with several micro- and great abscesses.
Microbiologic Analysis
We used sterile instruments to harvest soft tissue, tibia bone, and cement on day 14 (10 per group). The specimens of bone and soft tissues were added to 10 mL of phosphate-buffered saline and homogenized with a fast tissue grinder (JXFSTPRP-48). Then, 100 µL of supernatant was inoculated on Mueller-Hinton dishes and incubated for 24 hours at 37°C. The retrieved cement was placed in 2 mL of phosphate-buffered saline and sonicated to release bacterial biofilm from the cement [27]. A 100-µL aliquot of cement supernatant was plated in the same manner as the other tissue supernatants. Bacterial colonies were quantified using the plate count method [12].
Ethical Approval
This study was approved by the institutional review board at our institution. Ethical approval for this study was obtained from Wuhan University, Wuhan, China (number AF339).
Statistical Analysis
Data were analyzed using SPSS software (version 22.0, IBM Corp) and are presented as the mean ± standard error of the mean (means ± SEM). Normality was assessed using the Shapiro-Wilk test, with the Levene test used to determine the equality of variances. Independent-samples t-tests were used for between-group comparisons of normally distributed data. Within-group differences were analyzed with paired-samples t-tests or, where appropriate, repeated-measures ANOVA with a Dunnett post hoc analysis. We considered p values of < 0.05 to be significant.
Results
Mechanical Tests
All antibiotic-cement samples exceeded the compressive strength threshold of 70 MPa (requirement established in ISO 5833) [24] (p < 0.05) (Fig. 2D), but all samples had lower strength than unloaded cement samples. Of the antibiotic-loaded samples, aztreonam-loaded cement had better compressive properties before (p < 0.05) and after elution (p < 0.05).
Elution Assay
The release of the six antibiotic cements was rapid during the first day, especially piperacillin-loaded cement (p < 0.05) (Fig. 3A-D). The tobramycin-loaded cement maintained stable release at high concentrations within 28 days, and the major cumulative release proportion was 77% for 14 days, whereas the cumulative elution proportion of the other five tested antibiotic cements mainly occurred over the first 7 days (MEM 90%, PIP 99%, CAZ 94%, CIP 85%, ATM 93%), then continued more slowly throughout the study period. Piperacillin-loaded cement demonstrated the highest in vitro release among the six antibiotic cements in the elution kinetics within the first day and cumulative release over the first 7 days, whereas tobramycin cement showed the best release kinetics from day 3 to day 28 and cumulative release from day 7 to day 28 (p < 0.05) (Fig. 3A-D). The cumulative release proportions of the raw antibiotic in cement were 1182.8 ± 37.9 µg (TOB), 355.6 ± 16.2 µg (MEM), 721.2 ± 40.3 µg (PIP), 477.4 ± 37.1 µg (CAZ), 146.5 ± 11.3 µg (CIP), and 372.1 ± 14.5 µg (ATM) within 28 days (Table 1).
Fig. 3.
A-D This figure shows the mean elution kinetics and cumulative release of the tested antibiotic cement within 28 days. (A-B) These graphs show the mean elution kinetics of the tested antibiotic cement within 28 days. (C-D) These graphs show the mean cumulative release of the tested antibiotic cement within 28 days. The high-performance liquid chromatography–mass spectrometry parameter settings were: solvents: 1% formic acid water and pure acetonitrile; columns: Hypersil GOLD, Thermo Fisher, 100 × 2.1 mm 3 µm; flow rates: 0.2 mL/minute; time: 6 minute (TOB = filter: SRM MS2 163.21-468.33 m/z; mass: 163.21 m/z; retention time: 1.01 minute; MEM = filter: SRM MS2 141.15-384.18 m/z; mass: 141.15 m/z; retention time: 3.62 minute; PIP = filter: SRM MS2 143.13-518.26 m/z; mass: 143.13 m/z; retention time: 4.37 min; CAZ = filter: SRM MS2 468.15-547.22 m/z; mass: 468.15 m/z; retention time: 3.61 minute; CIP = filter: SRM MS2 288.28-332.23 m/z; mass: 288.28 m/z; retention time: 3.87 minute; ATM = filter: SRM MS2 96.03-434.14 m/z; mass: 96.03 m/z; retention time: 3.74 minute). The quantification limit for antibiotics was 0.1 µg/mL. There were six samples. ap < 0.01 (compared with the other antibiotic elution groups samples at the special time points).
Table 1.
Daily and cumulative release of antibiotics from cement specimens measured with elution assays over 28 days
| Release of antibiotics | Specimens | |||||
| TOB | MEM | PIP | CAZ | CIP | ATM | |
| Daily, µg/mL | ||||||
| Day 1 | 101.3 ± 6.1 | 53.2 ± 3.1 | 130.4 ± 8.1 | 81.8 ± 7.9 | 17.9 ± 1.8 | 59.2 ± 3.5 |
| Day 3 | 14.4 ± 2.2 | 5.5 ± 0.5 | 10.6 ± 2.1 | 4.9 ± 0.9 | 5.0 ± 0.7 | 5.8 ± 0.3 |
| Day 5 | 10.5 ± 2.0 | 2.4 ± 0.3 | 1.1 ± 0.1 | 1.8 ± 0.6 | 1.7 ± 0.3 | 2.6 ± 0.3 |
| Day 7 | 11.1 ± 1.8 | 2.1 ± 0.2 | 0.7 ± 0.2 | 1.7 ± 0.4 | 0.5 ± 0.3 | 1.7 ± 0.3 |
| Day 14 | 44.8 ± 3.2 | 3.1 ± 0.3 | 0.5 ± 0.0 | 1.8 ± 0.5 | 1.9 ± 0.2 | 2.5 ± 0.3 |
| Day 21 | 30.9 ± 5.3 | 2.7 ± 0.4 | 0.5 ± 0.0 | 1.7 ± 0.5 | 1.4 ± 0.3 | 1.6 ± 0.2 |
| Day 28 | 23.6 ± 4.8 | 2.1 ± 0.2 | 0.4 ± 0.0 | 1.6 ± 0.5 | 0.9 ± 0.3 | 1.0 ± 0.2 |
| Cumulative total, µg | 1182.8 ± 37.9 | 355.6 ± 16.2 | 721.2 ± 40.3 | 477.4 ± 37.1 | 146.5 ± 11.3 | 372.1 ± 14.5 |
All data are expressed as mean ± SEM.
Modified Microtube Dilution Bioassay of Elution Samples
During the 28-day study period, meropenem cement showed the best antimicrobial activity against the four tested bacteria (E. coli, P. aeruginosa, K. pneumoniae, and P. vulgaris) (Fig. 4A-D). In addition, ciprofloxacin cement inhibited E. coli growth, ceftazidime and aztreonam cement inhibited K. pneumoniae growth, and tobramycin cement inhibited P. aeruginosa growth (Table 2).
Fig. 4.
A-D These figures show antibacterial activities against tested gram-negative bacteria in broth elution samples from the antibiotic-loaded acrylic cement specimens over a 28-day elution period. The wells that appear cloudy indicate bacterial growth. The modified microtube dilation bioassay results of antibiotic activity of elution samples with (A) E. coli, (B) P. aeruginosa, (C) K. pneumoniae, and (D) P. vulgaris are shown. There were five samples.
Table 2.
Antibacterial efficacy as established using modified microtube dilution bioassay over 28 days
| Tested bacteria and cement specimens | Time after elution (days) | ||||||
| 1 | 3 | 5 | 7 | 14 | 21 | 28 | |
| E. coli ATCC 25922 | |||||||
| TOB | NG | + | + | + | NG | NG | + |
| MEM | NG | NG | NG | NG | NG | NG | NG |
| PIP | NG | NG | + | + | + | + | + |
| CAZ | NG | NG | NG | + | + | + | + |
| CIP | NG | NG | NG | NG | NG | NG | NG |
| ATM | NG | NG | NG | NG | NG | NG | + |
| P. aeruginosa ATCC 27853 | |||||||
| TOB | NG | NG | NG | NG | NG | NG | NG |
| MEM | NG | NG | NG | NG | NG | NG | NG |
| PIP | NG | + | + | + | + | + | + |
| CAZ | NG | + | + | + | + | + | + |
| CIP | NG | NG | NG | NG | NG | NG | + |
| ATM | NG | + | + | + | NG | NG | + |
| K. pneumoniae CMCC 46117 | |||||||
| TOB | NG | NG | + | + | NG | NG | + |
| MEM | NG | NG | NG | NG | NG | NG | NG |
| PIP | NG | + | + | + | + | + | + |
| CAZ | NG | NG | NG | NG | NG | NG | NG |
| CIP | NG | NG | + | + | + | + | + |
| ATM | NG | NG | NG | NG | NG | NG | NG |
| P. vulgaris CMCC 49027 | |||||||
| TOB | NG | + | + | + | NG | NG | + |
| MEM | NG | NG | NG | NG | NG | NG | NG |
| PIP | NG | + | + | + | + | + | + |
| CAZ | + | + | + | + | + | + | + |
| CIP | NG | NG | + | + | + | + | + |
| ATM | NG | + | + | + | + | + | + |
There were five eluted samples for each type antibiotic-loaded cement; NG = no bacterial growth; + = bacterial growth.
Kirby-Bauer Bioassay of Elution Samples
Meropenem cement was the only antibiotic that showed the best antimicrobial activity against the four tested bacteria (E. coli, P. aeruginosa, K. pneumoniae, and P. vulgaris) within 28 days (Fig. 5A-D), whereas the best antibacterial activity against K. pneumoniae was aztreonam cement (p < 0.01) (Fig. 5C). Moreover, ciprofloxacin cement also inhibited E. coli growth over 28 days (Fig. 5A).
Fig. 5.
A-D These figures show antibacterial activities against the tested gram-negative organisms by agar disk diffusion bioassay, as determined by the elution samples from the various antibiotic-loaded acrylic cement specimens over a 28-day elution period. The antimicrobial activities in elution samples with (A) E. coli, (B) P. aeruginosa, (C) K. pneumoniae, and (D) P. vulgaris treated with a Kirby-Bauer assay are shown. There were five samples. ap < 0.01 (compared with the other antibiotic elution group samples at the special time points). No antibacterial activity was set at 6 mm (the diameter of the PDM Diagnostic Disk), inhibition zone diameter of the released antibiotics was set as measured value minus 6 mm, and inhibition zones with a width 3 mm or less were scored as no antimicrobial activity; NS = no statistical significance.
Antimicrobial Effects of Meropenem Cement in a Rat Model
We observed no significant differences in body temperature among two groups throughout the study process (Fig. 6A). Compared with controls in the rat model, rats treated with meropenem cement showed improved body weight (control: 280.1 ± 4.2 g, MEM: 288.5 ± 6.6 g, mean difference 8.4 [95% CI: 4.3 to 12.6]; p < 0.001) (Fig. 6B), improved knee width (control: 13.5 ± 0.3 mm, MEM: 11.8 ± 0.4 mm, mean difference 1.7 [95% CI 1.4 to 2.0]; p < 0.001) (Fig. 6C), decreased inflammatory marker (control: 316.7 ± 45.0 mm, MEM: 116.5± 21.8 mm, mean difference 200.2 [95% CI 162.3 to 238.2]; p < 0.001) (Fig. 6D), decreased radiographic scores (control: 17.7 ± 2.0 mm, MEM: 10.7 ± 1.3 mm, mean difference 7.0 [95% CI 5.4 to 8.6]; p < 0.001) (Fig. 7A-B), improved BV/TV (control: 8.7 ± 3.0 mm, MEM: 28.5 ± 5.5 mm, mean difference 19.8 [95% CI 13.3 to 26.2]; p < 0.001) (Fig. 7A-C), decreased Rissing scale scores of the knee gross pathology (control: 3.3 ± 0.5, MEM: 1.1 ± 0.7, mean difference 2.2 [95% CI 1.7 to 2.7]; p < 0.001) (Fig. 7D-E), decreased Petty scale scores of knee synovium (control: 2.9 ± 0.4 mm, MEM: 0.7 ± 0.7 mm, mean difference 2.1 [95% CI 1.7 to 2.5]; p < 0.001) (Fig. 7D and 7F), and decreased bacterial counts of the bone and soft tissues and negative bacterial cultures of cement (p < 0.001, p < 0.001, p < 0.001, p < 0.001, respectively) (Fig. 8A-D).
Fig. 6.
A-D These graphs show changes in the general status and serum inflammatory markers of the rats. (A) This graph shows changes in body temperature during the study. An electric thermometer for animals and infrared thermometer were used to measure the rectal temperature of the rats preoperatively (day 0) and on days 1, 4, 7, and 14 postoperatively. There were 15 rats per group. (B) The rats’ weight was measured preoperatively (day 0) and on days 4, 7, and 14 postoperatively. There were 15 rats per group. (C) The maximum medial-lateral knee width of the rats was measured using precision electronic vernier calipers preoperatively (day 0) and on days 1, 4, 7, and 14 postoperatively. There were 15 rats per group. (D) The serum alpha-1-acid glycoprotein level of rats was measured preoperatively (day 0) and on days 7 and 14 postoperatively. There were six rats per group. ap < 0.01 (compared with the control [CON] group).
Fig. 7.
A-F These images show radiologic and histopathologic evaluations in the two groups at 14 days after surgery and bacterial inoculation. (A) Radiography (radiograph and micro-CT) was performed. (B) The mean radiographic score values were evaluated and analyzed. (C) Bone volume per trabecular volume (BV/TV) analysis of the proximal tibia of micro-CT. (D) A macroscopic examination of knee and tissue inflammation of the knee synovium with magnification (40× and 400×) were performed in each treatment group at 14 days postoperatively. (E) Modified Rissing scale score assessment on postoperative day 14 in each treatment group. (F) Mean knee synovium histological scores based on the criteria of a modified Petty scale on postoperative day 14 in each treatment group. There were 10 rats in each group, except for BV/TV, in which there were five. ap < 0.01 (compared with the control [CON] group).
Fig. 8.
A-D We analyzed microbial counts of tibia bone, soft tissues around the knee, cement, and the whole animal in each group. ap < 0.01 (compared with the control [CON] group).
Discussion
Although current studies mainly have focused on gram-positive bacteria such as Staphylococcus aureus and Staphylococcus epidermidis, skeletal infections caused by gram-negative bacteria do occur, accounting for 6% to 48% of all the pathogenic bacteria after primary arthroplasty operations [1, 2, 5, 11]. Even though some studies have evaluated a certain antibiotic against gram-negative bacteria-loaded acrylic cement, most studies have focused on one or two properties of antibiotic-loaded acrylic cement, such as mechanical properties, porosity, elution, or in vitro antimicrobial activities. No comparative studies between different antibiotic-loaded acrylic cements have been suggested to explore which antibiotics possessed with the best elution properties and/or antibacterial effects on the premise of maintaining mechanical properties. We therefore presented a comparative study for estimating different antibiotic-loaded acrylic cements and raised the questions: (1) Does the addition of antibiotics to acrylic cement adversely affect compressive strength before and after elution? (2) Which antibiotics have the highest daily concentration and cumulative release within 28 days? (3) Which antibiotics showed antimicrobial activity within 28 days? (4) Does antibiotic-loaded cement that showed stable antimicrobial activity in vitro exhibit the antibacterial effect in a rat model?
Limitations
Our study has limitations. First, although vacuum mixing was used clinically, all antibiotic-loaded cement samples in this study were prepared by manually mixing them under atmospheric conditions. We did this because vacuum mixing reduces antibiotic elution due to the decrease in cement porosity [19]. However, our results showed that the mechanical properties and cumulative release of antibiotic-loaded cement did not change with the preparation method under the condition of low-dose antibiotics compared with a previous study [24]. Second, we manually prepared the antibiotic-loaded cement cylinders by fully mixing the antibiotic with cement powder before the monomer was added; we did not use the method of suspending the antibiotic powder in the liquid monomer. However, antibiotic mixing methods have been demonstrated to have no effect on cumulative antibiotic elution [17]. Our results showed that the cumulative release of antibiotic-loaded cement did not decrease without using the suspending antibiotic method. Third, only a single acrylic cement formula was used in this study; the reason for choosing Simplex brand acrylic cement is that it is widely used in our country, and this is more meaningful for guiding our clinical work than other brands of acrylic cement. However, we should be cautious in extrapolating these results directly to other cement types. Fourth, although our results suggest that the compressive strength of ciprofloxacin-loaded acrylic cement was higher than 70 MPa, the compressive strength of wet samples soaked for 28 days increased compared with that of dry samples, and a previous study suggested that the dry sample of 5 wt% ciprofloxacin-loaded acrylic cement was higher than 70 MPa [24]. However, with the increase in soaking time, the compressive strength decreased and remained at 65 MPa on day 28 [8]. We considered that the possible reasons were caused by different brands of acrylic cement and different amounts of soaking liquid. Fifth, in our experiment, only male rats were selected for the in vivo animal tests, which may be because our previous studies were almost all conducted on male rats. In subsequent studies, animals of both sexes should be selected as much as possible to reduce the results bias caused by gender differences. Lastly, our animal study used meropenem-loaded acrylic cement without using tobramycin and other antibiotic-loaded cements for comparison. The reason was that our previous antibacterial experiment suggested that meropenem-loaded acrylic cement maintained stable antibacterial capability against four common gram-negative bacteria during the 28-day study period. Tobramycin-loaded acrylic cement, which exhibited the best in vitro cumulative release, may be considered for future in vivo studies to explore its antibacterial effects. Nevertheless, our in vitro antibacterial studies still have important clinical significance, suggesting that different antibiotic-loaded acrylic cements can be selected accordingly when targeting different gram-negative bacterial infection risks in the future.
Discussion of Key Findings
From the perspective of mechanical integrity of cement loaded with antibiotics against gram-negative organisms, our study indicated that using a threshold of 70 MPa, all tested antibiotic-loaded cements demonstrated acceptable compression strength before and after elution. Regarding the elution of antibiotics from the acrylic cement, tobramycin-loaded acrylic cement had the largest proportion released among the six tested antibiotic cements, maintaining stable release at high concentrations within 28 days, and the major cumulative release proportion was 77% during 14 days. For antibacterial properties, overall, meropenem eluted from the acrylic cement exhibited the best characteristics, showing activity against E. coli, P. aeruginosa, K. pneumoniae, and P. vulgaris for 28 days, whereas ciprofloxacin and aztreonam inhibited E. coli and K. pneumonia growth over 28 days, respectively. Regarding the in vivo performance of meropenem-loaded acrylic cement, when compared with controls (unloaded antibiotics) in the rat model, cement loaded with meropenem decreased inflammation and improved growth.
Conclusion
These results suggest that meropenem cement had the most stable in vitro antimicrobial activity and effective activity in a rat model, while also having acceptable mechanical and elution characteristics. A meropenem-loaded cement may be used as potentially effective prevention for skeletal infection caused by gram-negative bacteria.
Acknowledgments
We thank Yifan Sun MD and Ying Qi MD of the microbiology laboratory at the Affiliated Liutie Central Hospital of Guangxi Medical University of China for providing the test organisms: E. coli ATCC 25922, P. aeruginosa ATCC 27853, K. pneumoniae CMCC 46117, and P. vulgaris CMCC 49027. We thank Gaofu Wang MD and Xiping Wei MD of the Zhongnan Hospital of Wuhan University for their histopathologic evaluation of our animal study. We thank Hui Xie MD and Huijun Shen MD of Liuzhou People's Hospital for their radiological evaluation of our animal study. We thank Yinxian Wen for helping to design the research in the early stages of this study.
Footnotes
Each author certifies that there are no funding or commercial associations (consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article related to the author or any immediate family members.
All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on file with the publication and can be viewed on request.
Ethical approval for this study was obtained from Wuhan University, Wuhan, China (number AF339).
This work was performed at the Liuzhou People’s Hospital and Zhongnan Hospital of Wuhan University, China.
References
- 1.Abdel MP, Barreira P, Battenberg A, et al. hip and knee section, treatment, two-stage exchange spacer-related: Proceedings of International Consensus on Orthopedic Infections. J Arthroplasty. 2019;34:S427-S438. [DOI] [PubMed] [Google Scholar]
- 2.Benito N, Franco M, Ribera A, et al. Time trends in the aetiology of prosthetic joint infections: a multicentre cohort study. Clin Microbiol Iinfect. 2016;22:732 e731-e738. [DOI] [PubMed] [Google Scholar]
- 3.Bistolfi A, Ferracini R, Albanese C, et al. PMMA-based bone cements and the problem of joint arthroplasty infections: status and new perspectives. Materials (Basel). 2019;12:4002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chang Y, Tai CL, Hsieh PH, et al. Gentamicin in bone cement: a potentially more effective prophylactic measure of infection in joint arthroplasty. Bone Joint Res. 2013;2:220-226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cray C, Zaias J, Altman NH. Acute phase response in animals: a review. Comp Med. 2009;59:517-526. [PMC free article] [PubMed] [Google Scholar]
- 6.Delanois RE, Mistry JB, Gwam CU, et al. Current epidemiology of revision total knee arthroplasty in the United States. J Arthroplasty. 2017;32:2663-2668. [DOI] [PubMed] [Google Scholar]
- 7.Ensing GT, Van Horn JR, Van Der Mei HC, et al. Copal bone cement is more effective in preventing biofilm formation than Palacos R-G. Clin Orthop Relat Res. 2008;466:1492-1498. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gandomkarzadeh M, Moghimi HR, Mahboubi A. Evaluation of the effect of ciprofloxacin and vancomycin on mechanical properties of PMMA cement; a preliminary study on molecular weight. Sci Rep. 2020;10:3981. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hsieh PH, Lee MS, Hsu KY, et al. Gram-negative prosthetic joint infections: risk factors and outcome of treatment. Clin Infect Dis. 2009;49:1036-1043. [DOI] [PubMed] [Google Scholar]
- 10.Hsu YH, Hu CC, Hsieh PH, et al. Vancomycin and ceftazidime in bone cement as a potentially effective treatment for knee periprosthetic joint infection. J Bone Joint Surg Am. 2017;99:223-231. [DOI] [PubMed] [Google Scholar]
- 11.Kalbian IL, Goswami K, Tan TL, et al. Treatment outcomes and attrition in gram-negative periprosthetic joint infection. J Arthroplasty. 2020;35:849-854. [DOI] [PubMed] [Google Scholar]
- 12.Lovati AB, Bottagisio M, Maraldi S, et al. Vitamin E phosphate coating stimulates bone deposition in implant-related infections in a rat model. Clin Orthop Relat Res. 2018;476:1324-1338. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lovati AB, Drago L, Monti L, et al. Diabetic mouse model of orthopaedic implant-related Staphylococcus aureus infection. PLoS One. 2013;8:e67628. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lucke M, Schmidmaier G, Sadoni S, et al. Gentamicin coating of metallic implants reduces implant-related osteomyelitis in rats. Bone. 2003;32:521-531. [DOI] [PubMed] [Google Scholar]
- 15.Lucke M, Schmidmaier G, Sadoni S, et al. A new model of implant-related osteomyelitis in rats. J Biomed Mater Res B Appl Biomater. 2003;67:593-602. [DOI] [PubMed] [Google Scholar]
- 16.Matsumoto K, Nishi K, Kikuchi M, et al. Alpha1-acid glycoprotein suppresses rat acute inflammatory paw edema through the inhibition of neutrophils activation and prostaglandin E2 generation. Biol Pharm Bull. 2007;30:1226-1230. [DOI] [PubMed] [Google Scholar]
- 17.Mclaren AC, Nugent M, Economopoulos K, et al. Hand-mixed and premixed antibiotic-loaded bone cement have similar homogeneity. Clin Orthop Relat Res. 2009;467:1693-1698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mestriner FL, Spiller F, Laure HJ, et al. Acute-phase protein alpha-1-acid glycoprotein mediates neutrophil migration failure in sepsis by a nitric oxide-dependent mechanism. Proc Natl Acad Sci U S A. 2007;104:19595-19600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Meyer J, Piller G, Spiegel CA, et al. Vacuum-mixing significantly changes antibiotic elution characteristics of commercially available antibiotic-impregnated bone cements. J Bone Joint Surg Am. 2011;93:2049-2056. [DOI] [PubMed] [Google Scholar]
- 20.Miller RJ, Thompson JM, Zheng J, et al. In vivo bioluminescence imaging in a rabbit model of orthopaedic implant-associated infection to monitor efficacy of an antibiotic-releasing coating. J Bone Joint Surg Am. 2019;101:e12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Murillo O, Grau I, Lora-Tamayo J, et al. The changing epidemiology of bacteraemic osteoarticular infections in the early 21st century. Clin Microbiol Infect. 2015;21:254 e251-258. [DOI] [PubMed] [Google Scholar]
- 22.Rissing JP, Buxton TB, Weinstein RS, et al. Model of experimental chronic osteomyelitis in rats. Infect Immun. 1985;47:581-586. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Robinson DA, Bechtold JE, Carlson CS, et al. Development of a fracture osteomyelitis model in the rat femur. J Orthop Res. 2011;29:131-137. [DOI] [PubMed] [Google Scholar]
- 24.Slane J, Gietman B, Squire M. Antibiotic elution from acrylic bone cement loaded with high doses of tobramycin and vancomycin. J Orthop Res. 2018;36:1078-1085. [DOI] [PubMed] [Google Scholar]
- 25.Samuel S, Mathew BS, Veeraraghavan B, et al. In vitro study of elution kinetics and bio-activity of meropenem-loaded acrylic bone cement. J Orthop Traumatol. 2012;13:131-136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Thompson JM, Miller RJ, Ashbaugh AG, et al. Mouse model of gram-negative prosthetic joint infection reveals therapeutic targets. JCI Insight. 2018;3:e121737. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Thompson JM, Saini V, Ashbaugh AG, et al. Oral-only linezolid-rifampin is highly effective compared with other antibiotics for periprosthetic joint infection: study of a mouse model. J Bone Joint Surg Am. 2017;99:656-665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Tyas B, Marsh M, Oswald T, et al. Antibiotic resistance profiles of deep surgical site infections in hip hemiarthroplasty; comparing low dose single antibiotic versus high dose dual antibiotic impregnated cement. J Bone Jt Infect. 2018;3:123-129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Wei J, Tong K, Wang H, et al. Intra-articular vancomycin for the prophylaxis of periprosthetic joint infection caused by methicillin-resistant S. aureus after total knee arthroplasty in a rat model: the dosage, efficacy, and safety. Antimicrob Agents Chemother. 2021;66:e0164121. [DOI] [PMC free article] [PubMed] [Google Scholar]