Treatment of intramedullary osteomyelitis is difficult. The original report by Cierny and Mader1 described adequate surgical debridement, fracture stabilization, antibiotic administration, and soft tissue coverage.1–3 Antibiotic cement has been used in several forms to treat a variety of orthopedic infections since its first description in 1970 by Buchholz and Engelbrecht.4 Antibiotic-coated nails are frequently used as an adjunct treatment for infected long bones.5 This delivery method allows the use of high local doses of antibiotic with minimal risk of systemic effects. These infections are often associated with nonunion, making treatment difficult. There is a need for effective local antibiotic treatment along with implantation of a device to aid in stability.
Various forms of antibiotic-coated nails are used. Nails that are made of polymethylmethacrylate and have a guidewire are believed to provide effective elution but do not offer the stability that an interlocking nail can provide. Numerous techniques for intraoperative fabrication of antibiotic rods have been described, including use of a mold,5,6 manual rolling of the cement,7 and use of a chest tube as a mold.7–11 The practice of inserting the nail into a silicone tube filled with polymethylmethacrylate cement and allowing it to harden is superior to the use of commercially available antibiotic nail molds.6 Several studies have described elution of antibiotics from cement.12–16 These studies routinely showed synergistic release with combinations of antibiotics, a correlation between porosity and elution, and a high initial rate of elution that tapers over time. However, there has been no comparison of the use of antibiotic cement on intramedullary nails vs guidewires.
Application of vancomycin to an aminoglycoside-containing cement offers synergistic elution,12–14 probably as a result of the increased porosity induced by the additional passive soluble additive.13 Other literature showed that only 10% of the total antibiotic elutes from the cement because of hydrophobicity.17 Adding liquid gentamicin can substantially increase the porosity of the acrylic bone cement.18 Neut et al15 showed reduced release of gentamicin when the cement and the antibiotics were mixed under negative pressure.15 These studies added to the current understanding of the interaction between multiple antibiotics and offered techniques to improve elution.
Specific clinical factors must be considered when deciding between an antibiotic-coated interlocking nail construct and a guidewire construct, especially mechanical strength. Antibiotic release also must be considered. To compare the characteristics of antibiotic release between antibiotic-coated interlocking nails and guidewires, this study evaluated the elution rates of tobramycin and vancomycin from polymethylmethacrylate cement applied to the 2 different types of support. The amount and efficacy of the antibiotics released were tested, and the porosity of the cement and the effect of antibiotic dosage were evaluated. Finally, curing temperatures of the cement mantles were examined to determine the effect of temperature on the elution of antibiotics.
Materials and Methods
Interlocking nails coated with antibiotic-supplemented cement were constructed according to a hybrid of protocols based on techniques using silicone tubing by Thonse and Conway6 and No. 40 French chest tubes by Riel and Gladden.19 The current study used 8-mm interlocking tibial nails (Expert Tibial Nail; DePuy Synthes, Paoli, Pennsylvania) because these are the smallest nails that are widely commercially available. A No. 40 French polyvinyl chloride (PVC) chest tube (10-mm internal diameter) (Atrium Medical, Hudson, New Hampshire) was chosen as the size constraint because this type of tube is readily available in hospitals that treat patients with infected tibial nonunion and it is the largest size available. The cement was mixed in a cement mixer without suction (Revolution Cement Mixing System; Stryker Orthopaedics, Mahwah, New Jersey). Cement (40 g Simplex P containing 1 g tobramycin; Stryker) with and without additional antibiotic powder was mixed very well in the included mixing bowl before the monomer was added. The interlocking tibial nails were precut into 2-cm sections with a band saw and inserted into the chest tube. Two-sided tape was used to hold the interlocking nails together and keep cement off of the ends. This precut technique was used because preliminary testing showed that the 1-mm–thick coating of cement easily fractured off of the interlocking nails that were cut into segments after coating. Cement was then injected around the interlocking nail. Once the exothermic reaction began, the chest tube was scored. When the reaction was nearly finished, the chest tube was peeled off of the interlocking nail. Once the cement cooled, a band saw was used to cut it at the interface of each 2-cm section.
To make guidewires coated with antibiotic-supplemented cement, a 3.5-mm guidewire was placed into a No. 40 French chest tube. The tube was filled with antibiotic cement according to the protocol used for interlocking nails. The chest tube was scored again, and after the cement was nearly set, it was peeled away. These guidewires, with a 3.25-mm coating of cement, were cut into 2-cm sections with a band saw. The guidewires were cut after the nail was created because preliminary testing showed that it was difficult to make samples with consistent cement thickness on precut guidewires.
For both types of metal support, samples were divided into those with Simplex cement with 1 g tobramycin premixed as the low dose and those with Simplex cement with 1 g tobramycin premixed with the addition of 1 g vancomycin (Hospira, Lake Forest, Illinois) and an additional 1.2 g tobramycin (X-Gen Pharmaceuticals, Horseheads, New York) as the high dose. All additional antibiotics were added in powder form.
The study design is shown in Figure 1. Cement segments were placed into capped 15-mL Falcon polypropylene tubes (Becton Dickinson, Franklin Lakes, New Jersey) containing 3 mL sterile phosphate-buffered saline at room temperature. The segments were removed and placed in fresh phosphate-buffered saline at 15 and 30 minutes and at 1, 4, 8, 24, 48, 72, 168, 336, and 1008 hours. These eluted samples were stored at −20°C until the end of the 6-week sampling period. Tobramycin concentrations in the eluent were determined with fluorescence polarization immunoassay (Integra 800; Roche Diagnostics Corp, Indianapolis, Indiana). Vancomycin concentrations were determined by homogeneous enzyme immunoassay (Modular Analytics P; Roche Diagnostics Corp). Both measurements were performed by the clinical laboratory at Providence-Providence Park Hospital.
Microcomputed tomography (vivaCT 40; Scanco Medical AG, Brüttisellen, Switzerland) was used to measure the properties of the cement that could affect antibiotic release. At the end of the 6-week elution period, the samples were scanned at energy of 70 kV, intensity of 114 μA, and resolution of 15 μm. Analysis of the reconstructed image data was performed with a constant threshold and 3-dimensional techniques. Porosity was calculated as 1 minus the bone volume/total volume ratio.
Microprobe thermometers (Pro Digital thermometer with folding probe, #1488; Taylor Precision Products, Las Cruces, New Mexico) (temperature range, 16°F–302°F) were used to measure the exothermic reaction temperatures of each type of cement on each type of support. Temperature was measured through predrilled holes in the sides of a chest tube mold containing segments of interlocking nails or guidewires. Two probes were placed simultaneously on the surface of each nail. Temperatures were reported as mean±SD.
A semiquantitative bacterial growth assay was used to quantify the amount of eluted antibiotics in the phosphate-buffered saline samples, as described previously, with minor modifications.20 Methicillin-sensitive Staphylococcus aureus (CDC 587, #49230, American Type Culture Collection, Manassas, Virginia) was used because it was susceptible to both of the study antibiotics. One hundred microliters of each phosphate-buffered saline sample containing eluted antibiotics was added to 2 mL log phase growing S aureus in Luria-Bertani broth (#BP1427-500; Fisher Scientific, Fair Lawn, New Jersey) and cultured at 37°C for 4 hours. Absorbance was read at 600 nm (ABS600) on a spectrophotometer. Readings were standardized against a control sample of 100 μL phosphate-buffered saline without antibiotic added to S aureus growing in Luria-Bertani broth, which was considered 0% inhibition. After the ABS600 at the start of the assay was subtracted for each sample, percent inhibition was calculated as follows:
Student's t test was used to compare means for 2 groups, and 1-way analysis of variance with the Bonferroni post hoc test for multiple comparisons was used to compare means for more than 2 groups. P<.05 was considered statistically significant.
To determine elution profiles for each construct, 2-cm samples of interlocking nails or guidewires coated with tobramycin-supplemented cement were cultured in phosphate-buffered saline for a 6-week period. No difference was noted in the amount of tobramycin eluted between the 2 constructs (Figure 2A). However, when interlocking nails and guidewires were coated with vancomycin- and tobramycin-supplemented cement, elution of tobramycin (Figure 2B) and vancomycin (Figure 2C) was greater from the interlocking nails than from the guidewires. Regardless of the antibiotic content of the cement, most elution occurred within the first 48 hours. Elution of the antibiotics continued throughout the 6-week study for both vancomycin and tobramycin.
Cumulative release of antibiotics at various time points after cement formation. Release of tobramycin for constructs coated with cement containing 1 g tobramycin (A). Release of tobramycin (B) and vancomycin (C) for constructs coated with cement containing 2.2 g tobramycin plus 1 g vancomycin.
The rate of release of antibiotics at each time point is shown in the Table. Similar rates of tobramycin release were seen for interlocking nails and guidewires when the cement was prepared with the standard 1 g tobramycin. However, when the cement was supplemented with 1 g vancomycin plus 1.2 g tobramycin, the rate of release of both antibiotics from the interlocking nails was greater than that from the guidewires. Differences were statistically significant (P<.05–P<.001, Student's t test) at each time point.
Antibiotic Elution Rates From Interlocking Nail and Guidewire Cement Constructs
Aliquots of all samples taken at elution time of 15 minutes, 2 weeks, and 6 weeks were evaluated for antibiotic activity based on inhibition of the growth of S aureus. For interlocking nails coated with cement supplemented with 1 g tobramycin, inhibition of bacterial growth was dependent on the concentration (Figure 3A). At concentrations greater than 100 μg/mL, the bacteria appeared to be maximally inhibited. When both vancomycin (1 g) and tobramycin (2.2 g) were included in the cement, at most time points the eluents had sufficient antibiotic to be maximally inhibitory (Figures 3B–C). The exception was the guidewire at 2 weeks (Figures 3B–C).
Efficacy of antibiotic in the eluent for inhibiting bacterial growth at various time points after cement formation in constructs coated with cement containing 1 g tobramycin (A) and constructs coated with cement containing 2.2 g tobramycin (B) plus 1 g vancomycin (C).
Given the large difference found for the 2 constructs in the amount of antibiotic released when antibiotic powder was added to the cement, the effect of porosity was considered. Microcomputed tomography was used to determine the density of the cement. Because the metal from the interlocking nail or guidewire interferes with the analysis, portions of the cement fractured from the metal were used. As shown in Figure 4, porosity did not differ significantly for 3 samples. However, the porosity of the cement on the guidewire supplemented with 1 g vancomycin plus 1.2 g tobramycin was significantly greater than that for the other 3 cement types. A 1-way analysis of variance with multiple Bonferroni comparisons showed a statistically significant difference (P<.05). However, no direct correlation was found between porosity and the amount of antibiotic measured in the eluent for any construct or all constructs grouped together.
Microcomputed tomography measurement of antibiotic-supplemented cement on interlocking nail and guidewire constructs.
Cement Exothermic Temperature
To determine whether differences in the cement curing temperature correlated with antibiotic elution profiles, the surface temperature of the cement curing constructs was measured. As shown in Figure 5, a significant difference in curing temperature was found between interlocking nails and guidewires (P<.05). Although all constructs had the same initial temperature, regardless of antibiotic dose, the interlocking nails reached a much lower maximum temperature. This finding suggests that a lower curing temperature could allow a greater release of the added antibiotic powder from the cement. The possibility that the higher curing temperature of the guidewire construct resulted in a microlayer of PVC from the mold tubing on the surface of the cement was considered, but no such evidence was found by tactile, microscopic, or dye analysis.
Initial and maximum curing temperatures of cement on interlocking nail or guidewire constructs.
Antibiotic release from cement over time follows a predictable curve, with a large initial release followed by a slow taper over time. Intuitively, samples containing more antibiotic would release higher levels of antibiotic during an elution study. However, in the current study, the results were counterintuitive. The high-dose guidewire nails had a greater volume of antibiotic-containing cement, but they released less antibiotic. The higher exothermic temperatures reached during the creation of a guidewire containing antibiotic may have played a role in the decreased elution. The manufacturer of the PVC chest tubes stated that the tubes melt at 170°F, and although the temperatures recorded in the current study did not reach that threshold, unobserved microscopic melting on the cement may have occurred. Kim et al21 studied different fabrication techniques and evaluated the ability to make an ideal antibiotic nail and found that removing the chest tube from a guidewire nail that was allowed to cool by convection caused plastic to adhere to the nail. To avoid this problem, in the current study, the nails were not allowed to cool fully. However, no discernible plastic was observed with light microscopy or microcomputed tomography.
Both nails had efficacious antibiotic release for the 6-week study period, and the eluted samples inhibited bacterial growth, even at the 6-week interval. This finding suggests that cement from both the interlocking nails and the guidewires continued to release effective antibiotic for the entire 6-week study period. This is a common duration of intravenous antibiotic treatment for osteomyelitis. In addition, the antibacterial activity of vancomycin may have been affected by the confounding factor of the addition of tobramycin. For this reason, it is difficult to know at what concentration vancomycin by itself would have inhibited the growth of S aureus.
The data also showed that antibiotic-supplemented cement had greater antibiotic release from interlocking nails than from guidewires, although both released enough antibiotic to be maximally effective. The interlocking nails had a superior elution curve; however, the added cost of the implant may not be warranted when stability is not an issue. Additionally, removing an interlocking nail may be difficult because of potential delamination of the cement from the implant. Clinically, this is more common in interlocking nails vs guidewire nails because guidewire nails have a much thicker mantle.
This study had some limitations. Simplex cement was used rather than Palacos (Zimmer Biomet, Warsaw, Indiana) cement to allow comparison of results with the antibiotic elution literature and because it was more readily available. Studies have shown better antibiotic elution with Palacos cement compared with Simplex.22–26 Polyvinyl chloride chest tubes were chosen instead of silicone because they are commonly used at Detroit Receiving Hospital. The lower melting point of PVC compared with silicone potentially could confound the results. Silicone chest tubes also cost approximately $10 more per unit, which is 1 reason why they are not generally stocked at Detroit Receiving Hospital. Creating cement-coated interlocking nails or guidewires requires much experience, and removing the chest tube at the right time is an art. Inconsistencies in this procedure may have contributed to the heterogeneity of the results. This difficulty has been noted by other authors, and suggestions to improve the reproducibility of the construction of antibiotic nails have been published.21 In addition, it was necessary to cut the interlocking nails before they were coated in cement and to cut the guidewire nails after they were coated in cement, leaving the guidewire nails with a cut end. Because antibiotic release was similar for interlocking nails and guidewires coated with cement containing 1 g tobramycin, difference in cutting was not considered a confounding variable. Antibiotic release from the guidewires coated with cement containing 1 g vancomycin plus 2.2 g tobramycin was statistically lower than that from interlocking nails coated with these antibiotics, even with the added surface area. Hence, even with the added surface area, the guidewire nails had a lower elution curve than the interlocking nails. Finally, this study used an in vitro model, and an intramedullary canal is not a static environment. Continuous turnover of interstitial fluid and interaction of blood with the cement in vivo may result in different elution characteristics. Despite these limitations, the study results provide important information that can help in the selection of an implant for the treatment of infected long bone nonunion.
Interlocking nails and guidewires coated with antibiotic-supplemented cement have predictable elution profiles and provide efficacious antibiotic release for at least 6 weeks. These are viable treatment options for infected long bones. Further research is needed to determine the role of temperature in the surprising elution curves shown in this study.
- Cierny G III, Mader JT. Approach to adult osteomyelitis. Orthop Rev. 1987; 16(4):259–270.
- Patzakis MJ, Wilkins J, Wiss DA. Infection following intramedullary nailing of long bones: diagnosis and management. Clin Orthop Relat Res. 1986; 212:182–191.
- Patzakis MJ, Zalavras CG. Chronic posttraumatic osteomyelitis and infected nonunion of the tibia: current management concepts. J Am Acad Orthop Surg. 2005; 13(6):417–427. doi:10.5435/00124635-200510000-00006 [CrossRef]
- Buchholz HW, Engelbrecht H. Uber die Depotwirkung einiger Antibiotica bei Vermischung mit dem Kunstharz Palacos. Chirurg. 1970; 41:511–515.
- Thonse R, Conway J. Antibiotic cement-coated interlocking nail for the treatment of infected nonunions and segmental bone defects. J Orthop Trauma. 2007; 21(4):258–268. doi:10.1097/BOT.0b013e31803ea9e6 [CrossRef]
- Thonse R, Conway JD. Antibiotic cement-coated nails for the treatment of infected non-unions and segmental bone defects. J Bone Joint Surg Am. 2008; 90(suppl 4):163–174. doi:10.2106/JBJS.H.00753 [CrossRef]
- Shyam AK, Sancheti PK, Patel SK, Rocha S, Pradhan C, Patil A. Use of antibiotic cement-impregnated intramedullary nail in treatment of infected non-union of long bones. Indian J Orthop. 2009; 43(4):396–402. doi:10.4103/0019-5413.55468 [CrossRef]
- Wasko MK, Borens O. Antibiotic cement nail for the treatment of posttraumatic intramedullary infections of the tibia: midterm results in 10 cases. Injury. 2013; 44(8):1057–1060. doi:10.1016/j.injury.2013.05.001 [CrossRef]
- Paley D, Herzenberg JE. Intramedullary infections treated with antibiotic cement rods: preliminary results in nine cases. J Orthop Trauma. 2002; 16(10):723–729. doi:10.1097/00005131-200211000-00007 [CrossRef]
- Madanagopal SG, Seligson D, Roberts CS. The antibiotic cement nail for infection after tibial nailing. Orthopedics. 2004; 27(7):709–712.
- Qiang Z, Jun PZ, Jie XJ, Hang L, Bing LJ, Cai LF. Use of antibiotic cement rod to treat intramedullary infection after nailing: preliminary study in 19 patients. Arch Orthop Trauma Surg. 2007; 127(10):945–951. doi:10.1007/s00402-007-0315-x [CrossRef]
- Simpson PM, Dall GF, Breusch SJ, Heisel C. In vitro elution and mechanical properties of antibiotic-loaded SmartSet HV and Palacos R acrylic bone cements [in German]. Orthopade. 2005; 34(12):1255–1262. doi:10.1007/s00132-005-0861-2 [CrossRef]
- Penner MJ, Masri BA, Duncan CP. Elution characteristics of vancomycin and tobramycin combined in acrylic bone-cement. J Arthroplasty. 1996; 11(8):939–944. doi:10.1016/S0883-5403(96)80135-5 [CrossRef]
- Bridgens J, Davies S, Tilley L, Norman P, Stockley I. Orthopaedic bone cement: do we know what we are using?J Bone Joint Surg Br. 2008; 90(5):643–647. doi:10.1302/0301-620X.90B5.19803 [CrossRef]
- Neut D, van de Belt H, van Horn JR, van der Mei HC, Busscher HJ. The effect of mixing on gentamicin release from polymethylmethacrylate bone cements. Acta Orthop Scand. 2003; 74(6):670–676. doi:10.1080/00016470310018180 [CrossRef]
- Masri BA, Duncan CP, Beauchamp CP. Long-term elution of antibiotics from bone-cement: an in vivo study using the prosthesis of antibiotic-loaded acrylic cement (PROSTALAC) system. J Arthroplasty. 1998; 13(3):331–338. doi:10.1016/S0883-5403(98)90179-6 [CrossRef]
- DiCicco M, Duong T, Chu A, Jansen SA. Tobramycin and gentamicin elution analysis between two in situ polymerizable orthopedic composites. J Biomed Mater Res B Appl Biomater. 2003; 65(1):137–149. doi:10.1002/jbm.b.10528 [CrossRef]
- Hsieh PH, Tai CL, Lee PC, Chang YH. Liquid gentamicin and vancomycin in bone cement: a potentially more cost-effective regimen. J Arthroplasty. 2009; 24(1):125–130. doi:10.1016/j.arth.2008.01.131 [CrossRef]
- Riel RU, Gladden PB. A simple method for fashioning an antibiotic cement-coated interlocking intramedullary nail. Am J Orthop (Belle Mead NJ). 2010; 39(1):18–21.
- Song W, Ren W, Wan C, et al. A novel strontium-doped calcium polyphosphate/erythromycin/poly(vinyl alcohol) composite for bone tissue engineering. J Biomed Mater Res A. 2011; 98(3):359–371. doi:10.1002/jbm.a.33127 [CrossRef]
- Kim JW, Cuellar DO, Hao J, Seligson D, Mauffrey C. Custom-made antibiotic cement nails: a comparative study of different fabrication techniques. Injury. 2014; 45(8):1179–1184. doi:10.1016/j.injury.2014.03.006 [CrossRef]
- Hoff SF, Fitzgerald RH Jr, Kelly PJ. The depot administration of penicillin G and gentamicin in acrylic bone cement. J Bone Joint Surg Am. 1981; 63(5):798–804. doi:10.2106/00004623-198163050-00015 [CrossRef]
- Marks KE, Nelson CL, Lautenschlager EP. Antibiotic-impregnated acrylic bone cement. J Bone Joint Surg Am. 1976; 58(3):358–364. doi:10.2106/00004623-197658030-00011 [CrossRef]
- Wahlig H, Dingeldein E. Antibiotics and bone cements: experimental and clinical long-term observations. Acta Orthop Scand. 1980; 51(1):49–56. doi:10.3109/17453678008990768 [CrossRef]
- Elson RA, Jephcott AE, McGechie DB, Verettas D. Antibiotic-loaded acrylic cement. J Bone Joint Surg Br. 1977; 59(2):200–205.
- Berry DJ, Chandler HP, Reilly DT. The use of bone allografts in two-stage reconstruction after failure of hip replacements due to infection. J Bone Joint Surg Am. 1991; 73(10):1460–1468. doi:10.2106/00004623-199173100-00003 [CrossRef]
Antibiotic Elution Rates From Interlocking Nail and Guidewire Cement Constructs
|Time h||Mean±SD Antibiotic Content of Cement, µg/mL/h|
|1 g Tobramycin||2.2 g Tobramycin + 1 g Vancomycin|