- Email: [email protected]
Systemic Absorption of Antibiotics From Antibiotic-Loaded Cement Spacers for the Treatment of Periprosthetic Joint Infection
The Journal of Arthroplasty xxx (2017) 1e5
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Systemic Absorption of Antibiotics From Antibiotic-Loaded Cement Spacers for the Treatment of Periprosthetic Joint Infection Adam I. Edelstein, MD a, *, Kamil T. Okroj, MD b, Thea Rogers, MS c, Craig J. Della Valle, MD d, Scott M. Sporer, MD c, d a
Department of Orthopaedic Surgery, Northwestern University, Feinberg School of Medicine, Chicago, Illinois Department of Orthopaedic Surgery, Thomas Jefferson University, Philadelphia, Pennsylvania Northwestern Medicine Central DuPage Hospital, Joint Replacement Institute, Winfield, Illinois d Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, Illinois b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 July 2017 Received in revised form 14 September 2017 Accepted 23 September 2017 Available online xxx
Background: Two-stage treatment of periprosthetic joint infections involves placement of high-dose antibiotic-loaded cement spacers (ACSs). Reports of ACS-induced nephrotoxicity have raised concern regarding systemic absorption of antibiotics after ACS placement. We sought to characterize the serum concentrations of antibiotics that occur after ACS placement. Methods: We performed a prospective study of patients with an infected primary total hip (THA) or knee arthroplasty (TKA) treated with standardized ACSs with vancomycin, gentamicin, and tobramycin. Serum antibiotic levels were collected weekly for 8 weeks. Results: Twenty-one patients (10 THA, 11 TKA) were included. Mean serum gentamicin levels ranged between 0.275±0.046 and 0.364±0.163 mg/L; mean serum tobramycin levels ranged from 0.313±0.207 to 0.527±0.424 mg/L; and mean serum vancomycin levels ranged from 5.46±6.6 to 15.34±9.6 mg/L. Serum antibiotic levels were detectable throughout the 8-week duration of ACS treatment. Regression analysis found that diabetes (coefficient 6.73, 95% CI 0.92-12.54, P < .05), blood urea nitrogen (coefficient 0.83, 95% CI 0.45-1.22, P < .001), number of cement doses (coefficient 3.71, 95% CI 0.76-6.66, P < .05), and use of systemic vancomycin (coefficient 6.24, 95% CI 2.72-9.75, P < .001) correlated with serum vancomycin levels. Patient age (coefficient 0.01, 95% CI 0.02 to 0, P < .01) and male sex (coefficient 0.20, 95% CI 00.41, P < .05) correlated with serum aminoglycoside level. Conclusion: Systemic absorption of antibiotics from high-dose ACS persists for at least 8 weeks. Patients should be monitored closely for complications related to systemic absorption of antibiotics from ACS treatment. © 2017 Elsevier Inc. All rights reserved.
Keywords: antibiotic-loaded cement spacer periprosthetic joint infection serum levels systemic absorption systemic concentration
Despite the advent of modern antisepsis protocols in total joint arthroplasty (TJA), the projected increase in TJA utilization will yield over 250,000 hip and knee periprosthetic joint infections
One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to https://doi.org/10.1016/j.arth.2017.09.043. This study was completed after approval by the institutional review board. The work was performed at the Northwestern Medicine Central DuPage Hospital, Winfield, IL. * Reprint requests: Adam I. Edelstein, MD, 676 N. St. Clair Street, Suite 1350, Chicago, IL 60611. https://doi.org/10.1016/j.arth.2017.09.043 0883-5403/© 2017 Elsevier Inc. All rights reserved.
(PJIs) annually by the year 2030 [1]. The projected PJI burden demands the development of safe and efficacious treatment protocols for PJI. Two-stage treatment of PJI, consisting of joint resection and insertion of a high-dose antibiotic-loaded cement spacer (ACS) followed by delayed reimplantation of components, is the current gold standard with reported failure occurring in <10% of cases in some series [2,3]. The delay between joint resection and reimplantation of components is typically at least 6 weeks [4]. Studies of antibiotic concentrations achieved within the joint from ACS treatment have shown high local concentrations that are often many times above the minimum inhibitory concentration and persist throughout the duration of ACS treatment [5e8]. Previous studies of systemic absorption of antibiotics from cement have found that systemic concentrations peak several
2
A.I. Edelstein et al. / The Journal of Arthroplasty xxx (2017) 1e5
hours after surgery and rapidly decrease within 1-2 days [9e11]. Several reports on use of antibiotic-impregnated cement in TJA have documented low rates of systemic toxicity [12e14]. The Infectious Diseases Society of America stated in published guidelines on the treatment of PJI that systemic toxicity from ACS treatment is rare [4]. More recently, reports have emerged of serious systemic toxicity related to use of ACS including acute kidney injury [15]. Menge et al [16] reported a 17% rate of nephrotoxicity after the placement of an ACS for the treatment of knee PJI. The high incidence of systemic toxicity in these series contradicts earlier reports of low systemic absorption of antibiotics from cement spacers. Moreover, a report by Aeng et al found that in the week after surgery, 22% of patients treated with ACS for PJI had serum tobramycin concentrations >2 mg/L, a known threshold for the risk of nephrotoxicity [17]. Given the conflicting evidence regarding systemic toxicity from use of ACS, we sought to investigate the systemic antibiotic concentrations that occur after the placement of an ACS for the treatment of PJI.
Materials and Methods This study was a prospective study of patients with an infected primary hip or knee arthroplasty who underwent 2-stage treatment with joint resection and placement of an antibiotic-loaded polymethylmethacrylate spacer. Patients were recruited from a tertiary referral center between 2014 and 2016. ACSs were made using PALACOS RþG Bone Cement (Zimmer Biomet, Warsaw, IN), each pack of which contained 40 g of polymethylmethacrylate with 0.5 g of gentamicin. To each pack was admixed 3 g of vancomycin and 2.4 g of tobramycin. Cement spacers were made with between 2 and 4 packs of antibiotic cement; the exact amount used depended on patient size and the degree of bone loss. The decision to use a static or articulating spacer was based on the soft tissue status of the infected joint. After surgery, patients underwent treatment for 6 weeks with culture-directed intravenous (IV) antibiotics under the direction of an infectious disease specialist. Patients were then given a 2-week antibiotic holiday followed by second-stage revision if laboratory and aspiration results were consistent with infection eradication. We excluded patients with allergies to vancomycin or aminoglycosides. We also excluded patients who received parenteral aminoglycoside therapy to allow for measurement of serum aminoglycoside levels attributable only to the ACS. We included patients who received parenteral vancomycin therapy, but tracked this variable to discern the impact of vancomycin in the ACS irrespective of systemic therapy. Patients had weekly measurement of serum tobramycin, gentamycin, and vancomycin levels. For patients receiving systemic vancomycin, laboratory draws were performed as trough measurements. Patient characteristics including age, sex, race, body mass index, and diabetic status were documented. Mean serum concentrations of each antibiotic were calculated; vancomycin concentrations were calculated separately for the patients who did or did not receive systemic vancomycin treatment. At each weekly time point, the percentage of patients who had aminoglycoside or vancomycin levels above 2 mg/L or 15 mg/L, respectively, were calculated. These thresholds represent previously reported cutoffs above which the risk of nephrotoxicity increases [17,18]. In addition, regression models were fit to assess the relationship between patient characteristics, treatment variables, and the serum antibiotic levels. The level of significance was set at P < .05. Statistical analyses were performed using SAS statistical software (SAS Institute Inc, Cary, NC).
Table 1 Patient Characteristics. N ¼ 21
Demographics Sex Male Female Race White Other Age, mean (SD), y BMI, mean (SD), kg/m2 Location Knee Hip Type of spacer Static Articulating Type of cement PALACOS R&G Other Number of doses of cement, mean (SD) Discharged on IV antibiotics Vancomycin Daptomycin Penicillin G Ceftriaxone Cefazolin
13 (61.9) 8 (38.1) 18 3 65.6 32.2
(85.7) (14.3) (13.6) (10.8)
11 (52.4) 10 (47.6) 8 (38.1) 13 (61.9) 21 (100) 0 3.3 (0.6) 9 1 1 3 5
(42.9) (4.8) (4.8) (14.3) (23.8)
BMI, body mass index; IV, intravenous; SD, standard deviation.
Results We enrolled 22 patients in the study, and following application of exclusion criteria, the final analysis consisted of 21 patients (11 TKA and 10 THA). The mean age was 65.6 years and 62% were men. Eight received static spacers and 13 received articulating spacers. Patient characteristics are listed in Table 1. Mean weekly serum concentrations of gentamycin, tobramycin, and vancomycin were recorded for the 8-week duration of ACS treatment (Table 2 and Fig. 1). Mean serum gentamicin levels ranged between 0.275 ± 0.046 and 0.364 ± 0.163 mg/L; mean serum tobramycin levels ranged from 0.313 ± 0.207 to 0.527 ± 0.424 mg/L; and mean serum vancomycin levels ranged from 5.46 ± 6.6 to 15.34 ± 9.6 mg/L. Serum antibiotic levels were detectable throughout the 8-week duration of ACS treatment. Vancomycin levels were also recorded separately for those who did or did not receive systemic vancomycin (Table 3 and Fig. 2). Serum vancomycin levels were higher in patients treated with systemic vancomycin (range 8.35-21.5 mg/L), but remained detectable in those without systemic treatment (range 1.8-14.38 mg/L). The percentage of patients who reached nephrotoxicity threshold concentrations is reported in Table 4. No patients had recorded aminoglycoside levels above the nephrotoxicity threshold (2 mg/L) during the study. Depending on the week, between 14.3%
Table 2 Serum Antibiotic Concentrations.
Week Week Week Week Week Week Week Week
1 2 3 4 5 6 7 8
Vancomycin, Ug/mL, Mean (SD)
Tobramycin, Ug/mL, Mean (SD)
Gentamicin, Ug/mL, Mean (SD)
16.97 14.64 11.80 9.90 10.11 10.80 5.54 10.32
0.27 0.35 0.41 0.39 0.32 0.38 0.30 0.59
0.30 0.28 0.33 0.34 0.40 0.32 0.38 0.30
(3.6) (9.5) (8.0) (8.3) (7.9) (8.5) (6.5) (10.2)
SD, standard deviation.
(0.10) (0.08) (0.18) (0.14) (0.21) (0.15) (0.20) (0.58)
(0) (0.05) (0.14) (0.14) (0.15) (0.16) (0.13) (0)
A.I. Edelstein et al. / The Journal of Arthroplasty xxx (2017) 1e5
3
Fig. 1. Serum antibiotic levels over time.
and 70% of patients had serum vancomycin concentrations above the nephrotoxicity threshold (15 mg/L). Regression analysis identified correlations between patient and treatment variables and serum antibiotic concentrations (Table 5). Diabetes (coefficient 6.73, 95% CI 0.92-12.54, P < .05), blood urea nitrogen (coefficient 0.83, 95% CI 0.45-1.22, P < .001), number of cement doses (coefficient 3.71, 95% CI 0.76-6.66, P < .05), and use of systemic vancomycin (coefficient 6.24, 95% CI 2.72-9.75, P < .001) correlated with serum vancomycin levels. Patient age (coefficient 0.01, 95% CI 0.02 to 0, P < .01) and male sex (coefficient 0.20, 95% CI 0-0.41, P < .05) correlated with serum aminoglycoside level. Discussion To treat the growing burden of PJIs, effective treatments that minimize the risk of systemic complications are needed. Although the efficacy of high-dose ACSs has been established for the treatment of PJI, their safety profile remains to be fully elucidated [2,3]. Recent reports of ACS-related nephrotoxicity, including in our own institution, led us to aim to investigate the systemic absorption of antibiotics that occurs from ACS in patients undergoing 2-stage treatment for PJI throughout the duration of their treatment. We found that systemic absorption of antibiotics from ACS treatment persists for at least 8 weeks after ACS placement. ACS treatment of PJI achieves high intra-articular concentrations of antibiotics while preserving joint space [2,12]. The high intra-articular antibiotic concentrations can achieve efficacy against bacteria not susceptible to systemic antibiotics [9,19].
Table 3 Serum Vancomycin Concentrations Stratified by Use of IV Vancomycin.
Week Week Week Week Week Week Week Week
1 2 3 4 5 6 7 8
Discharged on Vancomycin, Mean (SD)
Not Discharged on Vancomycin, Mean (SD)
18.70 16.17 16.35 13.26 13.08 15.55 8.35 21.50
14.38 13.10 7.76 7.21 7.89 6.05 1.80 7.53
(3.45) (5.86) (4.58) (5.76) (6.24) (2.46) (6.66) (0)
IV, intravenous; SD, standard deviation.
(0.94) (11.34) (7.73) (8.61) (7.77) (8.64) (0.94) (8.04)
Intra-articular concentrations after ACS placement remain detectable until the time of second stage reimplantation [8]. In a study of 49 patients undergoing second stage revision for PJI, Masri et al [7] found that effective intra-articular concentrations of tobramycin were maintained for at least 4 months after ACS placement. There is conflicting literature regarding the extent and duration of systemic absorption of antibiotics from cement in hip and knee arthroplasties. The lack of consensus is based, at least in part, on variation in cement characteristics, amount of cement used, amount of antibiotic added, and whether the cement was used for component fixation or as a cement spacer. Several studies of antibiotic cement used for component fixation have reported an early peak in systemic absorption of antibiotics that then rapidly decreases [9,10,20,21]. Sterling et al [9] reported on use of tobramycin-impregnated cement (1 g tobramycin per 40 g cement pack) in 10 primary hip arthroplasty patients and found systemic levels of tobramycin peaked at 0.94 mg/L 3 hours after surgery and declined to <0.2 mg/L by 48 hours. Similarly, Pritchett and Bortel documented an early systemic peak of 1.1 mg/L tobramycin, which then decreased to undetectable serum concentrations by 48 hours after primary hip arthroplasty with tobramycin-impregnated cement (1.2 g per 40 g cement pack) in 15 patients [10]. In contrast, other studies have documented sustained systemic absorption of antibiotics from cement. Soto-Hall et al [22] prospectively followed up 10 patients who received tobramycinimpregnated cement (0.5 g per 40 g cement pack) in revision hip arthroplasty and found that serum levels of tobramycin remained above 0.5 mg/L at 2 weeks after surgery. In the setting of ACS treatment for PJI, more antibiotics are typically added to the cement and reported systemic levels have been higher. Aeng et al measured serum antibiotic levels for 7 days after ACS placement (3.6 g tobramycin and 1.5 g vancomycin per 40 g cement pack) in a cohort of 50 patients [23]. They found that the serum tobramycin concentration peaked on postoperative day 1, with 22% of patients reaching a serum concentration >2 mg/L; among the subset of 10 patients who developed acute kidney injury, the mean serum tobramycin level remained 0.8 mg/L on postoperative day 7 [23]. In contrast, they reported that mean serum vancomycin levels were below the limit of detection at all time points. Kalil et al [24] reported a serum tobramycin level of 0.9 mg/L on postoperative day 38 in a patient treated with an ACS (2 g tobramycin per 40 g cement pack) for PJI. To our knowledge, there have been no prospective studies that investigated the systemic
4
A.I. Edelstein et al. / The Journal of Arthroplasty xxx (2017) 1e5
Fig. 2. Serum vancomycin levels over time, stratified by discharge on intravenous (IV) vancomycin.
absorption of antibiotics from ACS throughout the duration of treatment. We found that serum levels of aminoglycoside and vancomycin remain elevated throughout the duration of treatment. Risk factors for antibiotic-induced nephrotoxicity include high serum concentrations [17,25], prolonged duration of therapy [26,27], and exposure to multiple nephrotoxic antibiotic agents [28,29]. The potential for cumulative nephrotoxicity from ongoing systemic absorption of antibiotics is highlighted by the cohort reported by Menge et al, in which of 84 patients treated with an ACS for knee PJI, 17% developed acute kidney injury with a peak in the average creatinine occurring 30 days after ACS placement. When considering exposure to nephrotoxic antibiotics from ACS treatment, several additional variables become relevant. Certain types of cement have been shown to elute antibiotics more rapidly and for longer duration; specifically, PALACOS has shown greater elution properties than Simplex [7,11]. The use of generic vs proprietary formulations of antibiotics has also been shown to affect elution characteristics based on differences in particle size and diffusion characteristics [30]. Use of multiple antibiotic agents in an ACS, in addition to potentially compounding the nephrotoxic risk via systemic exposure to multiple antibiotics, may also enhance the elution characteristics of the constituent antibiotics and therefore increase systemic levels [7,31,32]. We found that age, male sex, diabetic status, and elevated blood urea nitrogen correlated with serum antibiotic levels. Unsurprisingly, coadministration of IV vancomycin correlated with serum vancomycin levels. Previous studies of aminoglycosideimpregnated cement have found that antibiotic dose in the cement correlated with systemic levels [9,23,33]. In this study, although the amount of antibiotics added to each pack of cement was standardized, there was variation in the amount of cement used to create the spacers, resulting in differences in the total dose of antibiotic per ACS. We found that the serum vancomycin level correlated with the total amount of vancomycin in the ACS; the correlation between serum aminoglycoside and total amount of aminoglycoside in the ACS did not reach statistical significance. We
postulate that lower aminoglycoside dosing and differences in elution characteristics between aminoglycosides and vancomycin may have led to this discrepancy. Comparison of our results with those of the Aeng study [23] provides further insight into the impact of ACS antibiotic dose on serum levels. The study by Aeng et al found that 22% of patients had serum aminoglycoside levels greater than 2 mg/L in the first week after ACS placement. This contradicts our finding that no patients reached the aminoglycoside nephrotoxicity threshold. This discrepancy likely reflects measurement of serum levels within the first week vs at the end of the first week, in addition to a higher dose of tobramycin in the Aeng study (3.6 g/pack) compared with this study (2.4 g/pack). Conversely, the Aeng study found no patients who reached the vancomycin nephrotoxicity threshold, whereas we found up to 70% of patients over the threshold, including those who did not receive systemic vancomycin. Again, there were differences in the amount of vancomycin added to the cement, with those in the Aeng study receiving 1.5 g/pack, whereas patients in this study received 3 g/pack. These findings reveal that antibiotic dosing in the ACS is likely a key determinant of serum antibiotic levels, although this relationship only reached significance for vancomycin in this study. There are several limitations of this study that merit consideration. First, there was variation in the number of cement packs used, the type of ACS used (articulating vs static), and the joint treated (hip vs knee). Although this makes exact correlation between serum antibiotic concentrations and ACS antibiotic dose challenging, this variety reflects the clinical reality. Second, the relatively small number of patients in this study reduces our ability to precisely identify correlations between serum antibiotic levels and patient factors. Third, the findings in this study relate to a specific antibiotic-loaded cement formulation and may not be generalizable to all ACSs. Nonetheless, all patients received a standardized spacer protocol made with commonly used antibiotic agents and we feel this presents an ideal situation for analysis. Fourth, 43% of the patients in this cohort received IV
Table 4 Percentage of Patients With Serum Antibiotic Concentrations Above Published Thresholds for Nephrotoxicity.
2 mcg/mL Aminoglycoside, % 15 mcg/mL Vancomycin, %
Week 1
Week 2
Week 3
Week 4
Week 5
Week 6
Week 7
Week 8
0 70.0
0 64.3
0 35.3
0 33.3
0 35.7
0 50.0
0 14.3
0 40.0
A.I. Edelstein et al. / The Journal of Arthroplasty xxx (2017) 1e5 Table 5 Correlation Coefficients of Preoperative and Intraoperative Variables With Postoperative Serum Antibiotic Concentrations.
Age Male Diabetes BMI SCr BUN Doses cement Discharged on IV vancomycin
Vancomycin Trough*10
Tobramycin þ Gentamicin Trough*10
0.016 0.062 6.731 0.138 5.939 0.832 3.712 6.238
0.011 0.202 0.340 0.006 0.118 0.004 0.031 0.086
(0.106 to 0.139) (3.177 to 3.300) (0.923-12.540)* (0.032 to 0.308) (15.482 to 3.605) (0.447-1.216)*** (0.759-6.664)* (2.725-9.752)***
(0.019 (0.001 (0.023 (0.017 (0.465 (0.018 (0.160 (0.126
to to to to to to to to
0.003)** 0.405)* 0.703) 0.005) 0.701) 0.027) 0.223) 0.298)
*** P < .001; **P < .01; *P < .05. BMI, body mass index; BUN, blood urea nitrogen; IV, intravenous; SCr, serum creatinine.
vancomycin throughout the study period. We attempted to avoid confounding by controlling for this variable in our analyses. Finally, other studies of systemic absorption from antibiotic cement have found that levels peaked within the first 24 hours after placement. Our intent was to study systemic levels over the duration of ACS treatment, and as such we did not assay serum antibiotic levels in the immediate post-operative period and our data therefore underestimates peak levels of systemic absorption of antibiotics. In sum, use of high-dose ACSs for the treatment of PJI leads to elevated systemic antibiotic concentrations for at least 8 weeks after placement. Absorption of antibiotics from ACS treatment poses a theoretical risk of systemic toxicity, with an increasing body of evidence demonstrating ACS-related renal toxicity. Patients should be closely monitored for complications related to ACS treatment. Further investigation is needed to identify strategies to minimize the risk of systemic toxicity for this population. References [1] Kurtz SM, Ong KL, Schmier J, Mowat F, Saleh K, Dybvik E, et al. Future clinical and economic impact of revision total hip and knee arthroplasty. J Bone Joint Surg Am 2007;89(Suppl 3):144e51. [2] Cui Q, Mihalko WM, Shields JS, Ries M, Saleh KJ. Antibiotic-impregnated cement spacers for the treatment of infection associated with total hip or knee arthroplasty. J Bone Joint Surg Am 2007;89:871e82. [3] Voleti PB, Baldwin KD, Lee GC. Use of static or articulating spacers for infection following total knee arthroplasty: a systematic literature review. J Bone Joint Surg Am 2013;95:1594e9. [4] Osmon DR, Berbari EF, Berendt AR, Lew D, Zimmerli W, Steckelberg JM, et al. Diagnosis and management of prosthetic joint infection: clinical practice guidelines by the Infectious Diseases Society of America. Clin Infect Dis 2013;56:e1e25. [5] Hanssen AD, Osmon DR, Patel R. Local antibiotic delivery systems: where are we and where are we going? Clin Orthop Relat Res 2005:111e4. [6] Anagnostakos K, Wilmes P, Schmitt E, Kelm J. Elution of gentamicin and vancomycin from polymethylmethacrylate beads and hip spacers in vivo. Acta Orthop 2009;80:193e7. [7] 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;13:331e8. [8] Mutimer J, Gillespie G, Lovering AM, Porteus AJ. Measurements of in vivo intra-articular gentamicin levels from antibiotic loaded articulating spacers in revision total knee replacement. Knee 2009;16:39e41. [9] Sterling GJ, Crawford S, Potter JH, Koerbin G, Crawford R. The pharmacokinetics of Simplex-tobramycin bone cement. J Bone Joint Surg Br 2003;85:646e9.
5
[10] Pritchett JW, Bortel DT. Tobramycin-impregnated cement in total hip replacements. Orthop Rev 1992;21:577e9. [11] Marks KE, Nelson CL, Lautenschlager EP. Antibiotic-impregnated acrylic bone cement. J Bone Joint Surg Am 1976;58:358e64. [12] Wentworth SJ, Masri BA, Duncan CP, Southworth CB. Hip prosthesis of antibiotic-loaded acrylic cement for the treatment of infections following total hip arthroplasty. J Bone Joint Surg Am 2002;84-A(Suppl 2):123e8. [13] Springer BD, Lee GC, Osmon D, Haidukewych GJ, Hanssen AD, Jacofsky DJ. Systemic safety of high-dose antibiotic-loaded cement spacers after resection of an infected total knee arthroplasty. Clin Orthop Relat Res 2004: 47e51. [14] Koo KH, Yang JW, Cho SH, Song HR, Park HB, Ha YC, et al. Impregnation of vancomycin, gentamicin, and cefotaxime in a cement spacer for two-stage cementless reconstruction in infected total hip arthroplasty. J Arthroplasty 2001;16:882e92. [15] Luu A, Syed F, Raman G, Bhalla A, Muldoon E, Hadley S, et al. Two-stage arthroplasty for prosthetic joint infection: a systematic review of acute kidney injury, systemic toxicity and infection control. J Arthroplasty 2013;28: 1490e1498.e1. [16] Menge TJ, Koethe JR, Jenkins CA, Wright PW, Shinar AA, Miller GG, et al. Acute kidney injury after placement of an antibiotic-impregnated cement spacer during revision total knee arthroplasty. J Arthroplasty 2012;27: 1221e1227.e1e2. [17] McCormack JP, Jewesson PJ. A critical reevaluation of the “therapeutic range” of aminoglycosides. Clin Infect Dis 1992;14:320e39. [18] Jeffres MN, Isakow W, Doherty JA, Micek ST, Kollef MH. A retrospective analysis of possible renal toxicity associated with vancomycin in patients with health care-associated methicillin-resistant Staphylococcus aureus pneumonia. Clin Ther 2007;29:1107e15. [19] Cunha BA. Aminoglycosides: current role in antimicrobial therapy. Pharmacotherapy 1988;8:334e50. [20] Goodell JA, Flick AB, Hebert JC, Howe JG. Preparation and release characteristics of tobramycin-impregnated polymethylmethacrylate beads. Am J Hosp Pharm 1986;43:1454e61. [21] Wahlig H, Dingeldein E. Antibiotics and bone cements. Experimental and clinical long-term observations. Acta Orthop Scand 1980;51:49e56. [22] Soto-Hall R, Saenz L, Tavernetti R, Cabaud HE, Cochran TP. Tobramycin in bone cement. An in-depth analysis of wound, serum, and urine concentrations in patients undergoing total hip revision arthroplasty. Clin Orthop Relat Res 1983:60e4. [23] Aeng ES, Shalansky KF, Lau TT, Zalunardo N, Li G, Bowie WR, et al. Acute kidney injury with tobramycin-impregnated bone cement spacers in prosthetic joint infections. Ann Pharmacother 2015;49:1207e13. [24] Kalil GZ, Ernst EJ, Johnson SJ, Johannsson B, Polgreen PM, Bertolatus JA, et al. Systemic exposure to aminoglycosides following knee and hip arthroplasty with aminoglycoside-loaded bone cement implants. Ann Pharmacother 2012;46:929e34. [25] Raveh D, Kopyt M, Hite Y, Rudensky B, Sonnenblick M, Yinnon AM. Risk factors for nephrotoxicity in elderly patients receiving once-daily aminoglycosides. QJM 2002;95:291e7. [26] Dahlgren JG, Anderson ET, Hewitt WL. Gentamicin blood levels: a guide to nephrotoxicity. Antimicrob Agents Chemother 1975;8:58e62. [27] Paterson DL, Robson JM, Wagener MM. Risk factors for toxicity in elderly patients given aminoglycosides once daily. J Gen Intern Med 1998;13:735e9. [28] Rybak MJ, Albrecht LM, Boike SC, Chandrasekar PH. Nephrotoxicity of vancomycin, alone and with an aminoglycoside. J Antimicrob Chemother 1990;25:679e87. [29] Bertino Jr JS, Booker LA, Franck PA, Jenkins PL, Franck KR, Nafziger AN. Incidence of and significant risk factors for aminoglycoside-associated nephrotoxicity in patients dosed by using individualized pharmacokinetic monitoring. J Infect Dis 1993;167:173e9. [30] McLaren RL, McLaren AC, Vernon BL. Generic tobramycin elutes from bone cement faster than proprietary tobramycin. Clin Orthop Relat Res 2008;466: 1372e6. [31] Gonzalez Della Valle A, Bostrom M, Brause B, Harney C, Salvati EA. Effective bactericidal activity of tobramycin and vancomycin eluted from acrylic bone cement. Acta Orthop Scand 2001;72:237e40. [32] Penner MJ, Masri BA, Duncan CP. Elution characteristics of vancomycin and tobramycin combined in acrylic bone-cement. J Arthroplasty 1996;11: 939e44. [33] Wahlig H, Dingeldein E, Buchholz HW, Buchholz M, Bachmann F. Pharmacokinetic study of gentamicin-loaded cement in total hip replacements. Comparative effects of varying dosage. J Bone Joint Surg Br 1984;66: 175e9.
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Systemic Absorption of Antibiotics From Antibiotic-Loaded Cement Spacers for the Treatment of Periprosthetic Joint Infection Adam I. Edelstein, MD a, *, Kamil T. Okroj, MD b, Thea Rogers, MS c, Craig J. Della Valle, MD d, Scott M. Sporer, MD c, d a
Department of Orthopaedic Surgery, Northwestern University, Feinberg School of Medicine, Chicago, Illinois Department of Orthopaedic Surgery, Thomas Jefferson University, Philadelphia, Pennsylvania Northwestern Medicine Central DuPage Hospital, Joint Replacement Institute, Winfield, Illinois d Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, Illinois b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 July 2017 Received in revised form 14 September 2017 Accepted 23 September 2017 Available online xxx
Background: Two-stage treatment of periprosthetic joint infections involves placement of high-dose antibiotic-loaded cement spacers (ACSs). Reports of ACS-induced nephrotoxicity have raised concern regarding systemic absorption of antibiotics after ACS placement. We sought to characterize the serum concentrations of antibiotics that occur after ACS placement. Methods: We performed a prospective study of patients with an infected primary total hip (THA) or knee arthroplasty (TKA) treated with standardized ACSs with vancomycin, gentamicin, and tobramycin. Serum antibiotic levels were collected weekly for 8 weeks. Results: Twenty-one patients (10 THA, 11 TKA) were included. Mean serum gentamicin levels ranged between 0.275±0.046 and 0.364±0.163 mg/L; mean serum tobramycin levels ranged from 0.313±0.207 to 0.527±0.424 mg/L; and mean serum vancomycin levels ranged from 5.46±6.6 to 15.34±9.6 mg/L. Serum antibiotic levels were detectable throughout the 8-week duration of ACS treatment. Regression analysis found that diabetes (coefficient 6.73, 95% CI 0.92-12.54, P < .05), blood urea nitrogen (coefficient 0.83, 95% CI 0.45-1.22, P < .001), number of cement doses (coefficient 3.71, 95% CI 0.76-6.66, P < .05), and use of systemic vancomycin (coefficient 6.24, 95% CI 2.72-9.75, P < .001) correlated with serum vancomycin levels. Patient age (coefficient 0.01, 95% CI 0.02 to 0, P < .01) and male sex (coefficient 0.20, 95% CI 00.41, P < .05) correlated with serum aminoglycoside level. Conclusion: Systemic absorption of antibiotics from high-dose ACS persists for at least 8 weeks. Patients should be monitored closely for complications related to systemic absorption of antibiotics from ACS treatment. © 2017 Elsevier Inc. All rights reserved.
Keywords: antibiotic-loaded cement spacer periprosthetic joint infection serum levels systemic absorption systemic concentration
Despite the advent of modern antisepsis protocols in total joint arthroplasty (TJA), the projected increase in TJA utilization will yield over 250,000 hip and knee periprosthetic joint infections
One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to https://doi.org/10.1016/j.arth.2017.09.043. This study was completed after approval by the institutional review board. The work was performed at the Northwestern Medicine Central DuPage Hospital, Winfield, IL. * Reprint requests: Adam I. Edelstein, MD, 676 N. St. Clair Street, Suite 1350, Chicago, IL 60611. https://doi.org/10.1016/j.arth.2017.09.043 0883-5403/© 2017 Elsevier Inc. All rights reserved.
(PJIs) annually by the year 2030 [1]. The projected PJI burden demands the development of safe and efficacious treatment protocols for PJI. Two-stage treatment of PJI, consisting of joint resection and insertion of a high-dose antibiotic-loaded cement spacer (ACS) followed by delayed reimplantation of components, is the current gold standard with reported failure occurring in <10% of cases in some series [2,3]. The delay between joint resection and reimplantation of components is typically at least 6 weeks [4]. Studies of antibiotic concentrations achieved within the joint from ACS treatment have shown high local concentrations that are often many times above the minimum inhibitory concentration and persist throughout the duration of ACS treatment [5e8]. Previous studies of systemic absorption of antibiotics from cement have found that systemic concentrations peak several
2
A.I. Edelstein et al. / The Journal of Arthroplasty xxx (2017) 1e5
hours after surgery and rapidly decrease within 1-2 days [9e11]. Several reports on use of antibiotic-impregnated cement in TJA have documented low rates of systemic toxicity [12e14]. The Infectious Diseases Society of America stated in published guidelines on the treatment of PJI that systemic toxicity from ACS treatment is rare [4]. More recently, reports have emerged of serious systemic toxicity related to use of ACS including acute kidney injury [15]. Menge et al [16] reported a 17% rate of nephrotoxicity after the placement of an ACS for the treatment of knee PJI. The high incidence of systemic toxicity in these series contradicts earlier reports of low systemic absorption of antibiotics from cement spacers. Moreover, a report by Aeng et al found that in the week after surgery, 22% of patients treated with ACS for PJI had serum tobramycin concentrations >2 mg/L, a known threshold for the risk of nephrotoxicity [17]. Given the conflicting evidence regarding systemic toxicity from use of ACS, we sought to investigate the systemic antibiotic concentrations that occur after the placement of an ACS for the treatment of PJI.
Materials and Methods This study was a prospective study of patients with an infected primary hip or knee arthroplasty who underwent 2-stage treatment with joint resection and placement of an antibiotic-loaded polymethylmethacrylate spacer. Patients were recruited from a tertiary referral center between 2014 and 2016. ACSs were made using PALACOS RþG Bone Cement (Zimmer Biomet, Warsaw, IN), each pack of which contained 40 g of polymethylmethacrylate with 0.5 g of gentamicin. To each pack was admixed 3 g of vancomycin and 2.4 g of tobramycin. Cement spacers were made with between 2 and 4 packs of antibiotic cement; the exact amount used depended on patient size and the degree of bone loss. The decision to use a static or articulating spacer was based on the soft tissue status of the infected joint. After surgery, patients underwent treatment for 6 weeks with culture-directed intravenous (IV) antibiotics under the direction of an infectious disease specialist. Patients were then given a 2-week antibiotic holiday followed by second-stage revision if laboratory and aspiration results were consistent with infection eradication. We excluded patients with allergies to vancomycin or aminoglycosides. We also excluded patients who received parenteral aminoglycoside therapy to allow for measurement of serum aminoglycoside levels attributable only to the ACS. We included patients who received parenteral vancomycin therapy, but tracked this variable to discern the impact of vancomycin in the ACS irrespective of systemic therapy. Patients had weekly measurement of serum tobramycin, gentamycin, and vancomycin levels. For patients receiving systemic vancomycin, laboratory draws were performed as trough measurements. Patient characteristics including age, sex, race, body mass index, and diabetic status were documented. Mean serum concentrations of each antibiotic were calculated; vancomycin concentrations were calculated separately for the patients who did or did not receive systemic vancomycin treatment. At each weekly time point, the percentage of patients who had aminoglycoside or vancomycin levels above 2 mg/L or 15 mg/L, respectively, were calculated. These thresholds represent previously reported cutoffs above which the risk of nephrotoxicity increases [17,18]. In addition, regression models were fit to assess the relationship between patient characteristics, treatment variables, and the serum antibiotic levels. The level of significance was set at P < .05. Statistical analyses were performed using SAS statistical software (SAS Institute Inc, Cary, NC).
Table 1 Patient Characteristics. N ¼ 21
Demographics Sex Male Female Race White Other Age, mean (SD), y BMI, mean (SD), kg/m2 Location Knee Hip Type of spacer Static Articulating Type of cement PALACOS R&G Other Number of doses of cement, mean (SD) Discharged on IV antibiotics Vancomycin Daptomycin Penicillin G Ceftriaxone Cefazolin
13 (61.9) 8 (38.1) 18 3 65.6 32.2
(85.7) (14.3) (13.6) (10.8)
11 (52.4) 10 (47.6) 8 (38.1) 13 (61.9) 21 (100) 0 3.3 (0.6) 9 1 1 3 5
(42.9) (4.8) (4.8) (14.3) (23.8)
BMI, body mass index; IV, intravenous; SD, standard deviation.
Results We enrolled 22 patients in the study, and following application of exclusion criteria, the final analysis consisted of 21 patients (11 TKA and 10 THA). The mean age was 65.6 years and 62% were men. Eight received static spacers and 13 received articulating spacers. Patient characteristics are listed in Table 1. Mean weekly serum concentrations of gentamycin, tobramycin, and vancomycin were recorded for the 8-week duration of ACS treatment (Table 2 and Fig. 1). Mean serum gentamicin levels ranged between 0.275 ± 0.046 and 0.364 ± 0.163 mg/L; mean serum tobramycin levels ranged from 0.313 ± 0.207 to 0.527 ± 0.424 mg/L; and mean serum vancomycin levels ranged from 5.46 ± 6.6 to 15.34 ± 9.6 mg/L. Serum antibiotic levels were detectable throughout the 8-week duration of ACS treatment. Vancomycin levels were also recorded separately for those who did or did not receive systemic vancomycin (Table 3 and Fig. 2). Serum vancomycin levels were higher in patients treated with systemic vancomycin (range 8.35-21.5 mg/L), but remained detectable in those without systemic treatment (range 1.8-14.38 mg/L). The percentage of patients who reached nephrotoxicity threshold concentrations is reported in Table 4. No patients had recorded aminoglycoside levels above the nephrotoxicity threshold (2 mg/L) during the study. Depending on the week, between 14.3%
Table 2 Serum Antibiotic Concentrations.
Week Week Week Week Week Week Week Week
1 2 3 4 5 6 7 8
Vancomycin, Ug/mL, Mean (SD)
Tobramycin, Ug/mL, Mean (SD)
Gentamicin, Ug/mL, Mean (SD)
16.97 14.64 11.80 9.90 10.11 10.80 5.54 10.32
0.27 0.35 0.41 0.39 0.32 0.38 0.30 0.59
0.30 0.28 0.33 0.34 0.40 0.32 0.38 0.30
(3.6) (9.5) (8.0) (8.3) (7.9) (8.5) (6.5) (10.2)
SD, standard deviation.
(0.10) (0.08) (0.18) (0.14) (0.21) (0.15) (0.20) (0.58)
(0) (0.05) (0.14) (0.14) (0.15) (0.16) (0.13) (0)
A.I. Edelstein et al. / The Journal of Arthroplasty xxx (2017) 1e5
3
Fig. 1. Serum antibiotic levels over time.
and 70% of patients had serum vancomycin concentrations above the nephrotoxicity threshold (15 mg/L). Regression analysis identified correlations between patient and treatment variables and serum antibiotic concentrations (Table 5). Diabetes (coefficient 6.73, 95% CI 0.92-12.54, P < .05), blood urea nitrogen (coefficient 0.83, 95% CI 0.45-1.22, P < .001), number of cement doses (coefficient 3.71, 95% CI 0.76-6.66, P < .05), and use of systemic vancomycin (coefficient 6.24, 95% CI 2.72-9.75, P < .001) correlated with serum vancomycin levels. Patient age (coefficient 0.01, 95% CI 0.02 to 0, P < .01) and male sex (coefficient 0.20, 95% CI 0-0.41, P < .05) correlated with serum aminoglycoside level. Discussion To treat the growing burden of PJIs, effective treatments that minimize the risk of systemic complications are needed. Although the efficacy of high-dose ACSs has been established for the treatment of PJI, their safety profile remains to be fully elucidated [2,3]. Recent reports of ACS-related nephrotoxicity, including in our own institution, led us to aim to investigate the systemic absorption of antibiotics that occurs from ACS in patients undergoing 2-stage treatment for PJI throughout the duration of their treatment. We found that systemic absorption of antibiotics from ACS treatment persists for at least 8 weeks after ACS placement. ACS treatment of PJI achieves high intra-articular concentrations of antibiotics while preserving joint space [2,12]. The high intra-articular antibiotic concentrations can achieve efficacy against bacteria not susceptible to systemic antibiotics [9,19].
Table 3 Serum Vancomycin Concentrations Stratified by Use of IV Vancomycin.
Week Week Week Week Week Week Week Week
1 2 3 4 5 6 7 8
Discharged on Vancomycin, Mean (SD)
Not Discharged on Vancomycin, Mean (SD)
18.70 16.17 16.35 13.26 13.08 15.55 8.35 21.50
14.38 13.10 7.76 7.21 7.89 6.05 1.80 7.53
(3.45) (5.86) (4.58) (5.76) (6.24) (2.46) (6.66) (0)
IV, intravenous; SD, standard deviation.
(0.94) (11.34) (7.73) (8.61) (7.77) (8.64) (0.94) (8.04)
Intra-articular concentrations after ACS placement remain detectable until the time of second stage reimplantation [8]. In a study of 49 patients undergoing second stage revision for PJI, Masri et al [7] found that effective intra-articular concentrations of tobramycin were maintained for at least 4 months after ACS placement. There is conflicting literature regarding the extent and duration of systemic absorption of antibiotics from cement in hip and knee arthroplasties. The lack of consensus is based, at least in part, on variation in cement characteristics, amount of cement used, amount of antibiotic added, and whether the cement was used for component fixation or as a cement spacer. Several studies of antibiotic cement used for component fixation have reported an early peak in systemic absorption of antibiotics that then rapidly decreases [9,10,20,21]. Sterling et al [9] reported on use of tobramycin-impregnated cement (1 g tobramycin per 40 g cement pack) in 10 primary hip arthroplasty patients and found systemic levels of tobramycin peaked at 0.94 mg/L 3 hours after surgery and declined to <0.2 mg/L by 48 hours. Similarly, Pritchett and Bortel documented an early systemic peak of 1.1 mg/L tobramycin, which then decreased to undetectable serum concentrations by 48 hours after primary hip arthroplasty with tobramycin-impregnated cement (1.2 g per 40 g cement pack) in 15 patients [10]. In contrast, other studies have documented sustained systemic absorption of antibiotics from cement. Soto-Hall et al [22] prospectively followed up 10 patients who received tobramycinimpregnated cement (0.5 g per 40 g cement pack) in revision hip arthroplasty and found that serum levels of tobramycin remained above 0.5 mg/L at 2 weeks after surgery. In the setting of ACS treatment for PJI, more antibiotics are typically added to the cement and reported systemic levels have been higher. Aeng et al measured serum antibiotic levels for 7 days after ACS placement (3.6 g tobramycin and 1.5 g vancomycin per 40 g cement pack) in a cohort of 50 patients [23]. They found that the serum tobramycin concentration peaked on postoperative day 1, with 22% of patients reaching a serum concentration >2 mg/L; among the subset of 10 patients who developed acute kidney injury, the mean serum tobramycin level remained 0.8 mg/L on postoperative day 7 [23]. In contrast, they reported that mean serum vancomycin levels were below the limit of detection at all time points. Kalil et al [24] reported a serum tobramycin level of 0.9 mg/L on postoperative day 38 in a patient treated with an ACS (2 g tobramycin per 40 g cement pack) for PJI. To our knowledge, there have been no prospective studies that investigated the systemic
4
A.I. Edelstein et al. / The Journal of Arthroplasty xxx (2017) 1e5
Fig. 2. Serum vancomycin levels over time, stratified by discharge on intravenous (IV) vancomycin.
absorption of antibiotics from ACS throughout the duration of treatment. We found that serum levels of aminoglycoside and vancomycin remain elevated throughout the duration of treatment. Risk factors for antibiotic-induced nephrotoxicity include high serum concentrations [17,25], prolonged duration of therapy [26,27], and exposure to multiple nephrotoxic antibiotic agents [28,29]. The potential for cumulative nephrotoxicity from ongoing systemic absorption of antibiotics is highlighted by the cohort reported by Menge et al, in which of 84 patients treated with an ACS for knee PJI, 17% developed acute kidney injury with a peak in the average creatinine occurring 30 days after ACS placement. When considering exposure to nephrotoxic antibiotics from ACS treatment, several additional variables become relevant. Certain types of cement have been shown to elute antibiotics more rapidly and for longer duration; specifically, PALACOS has shown greater elution properties than Simplex [7,11]. The use of generic vs proprietary formulations of antibiotics has also been shown to affect elution characteristics based on differences in particle size and diffusion characteristics [30]. Use of multiple antibiotic agents in an ACS, in addition to potentially compounding the nephrotoxic risk via systemic exposure to multiple antibiotics, may also enhance the elution characteristics of the constituent antibiotics and therefore increase systemic levels [7,31,32]. We found that age, male sex, diabetic status, and elevated blood urea nitrogen correlated with serum antibiotic levels. Unsurprisingly, coadministration of IV vancomycin correlated with serum vancomycin levels. Previous studies of aminoglycosideimpregnated cement have found that antibiotic dose in the cement correlated with systemic levels [9,23,33]. In this study, although the amount of antibiotics added to each pack of cement was standardized, there was variation in the amount of cement used to create the spacers, resulting in differences in the total dose of antibiotic per ACS. We found that the serum vancomycin level correlated with the total amount of vancomycin in the ACS; the correlation between serum aminoglycoside and total amount of aminoglycoside in the ACS did not reach statistical significance. We
postulate that lower aminoglycoside dosing and differences in elution characteristics between aminoglycosides and vancomycin may have led to this discrepancy. Comparison of our results with those of the Aeng study [23] provides further insight into the impact of ACS antibiotic dose on serum levels. The study by Aeng et al found that 22% of patients had serum aminoglycoside levels greater than 2 mg/L in the first week after ACS placement. This contradicts our finding that no patients reached the aminoglycoside nephrotoxicity threshold. This discrepancy likely reflects measurement of serum levels within the first week vs at the end of the first week, in addition to a higher dose of tobramycin in the Aeng study (3.6 g/pack) compared with this study (2.4 g/pack). Conversely, the Aeng study found no patients who reached the vancomycin nephrotoxicity threshold, whereas we found up to 70% of patients over the threshold, including those who did not receive systemic vancomycin. Again, there were differences in the amount of vancomycin added to the cement, with those in the Aeng study receiving 1.5 g/pack, whereas patients in this study received 3 g/pack. These findings reveal that antibiotic dosing in the ACS is likely a key determinant of serum antibiotic levels, although this relationship only reached significance for vancomycin in this study. There are several limitations of this study that merit consideration. First, there was variation in the number of cement packs used, the type of ACS used (articulating vs static), and the joint treated (hip vs knee). Although this makes exact correlation between serum antibiotic concentrations and ACS antibiotic dose challenging, this variety reflects the clinical reality. Second, the relatively small number of patients in this study reduces our ability to precisely identify correlations between serum antibiotic levels and patient factors. Third, the findings in this study relate to a specific antibiotic-loaded cement formulation and may not be generalizable to all ACSs. Nonetheless, all patients received a standardized spacer protocol made with commonly used antibiotic agents and we feel this presents an ideal situation for analysis. Fourth, 43% of the patients in this cohort received IV
Table 4 Percentage of Patients With Serum Antibiotic Concentrations Above Published Thresholds for Nephrotoxicity.
2 mcg/mL Aminoglycoside, % 15 mcg/mL Vancomycin, %
Week 1
Week 2
Week 3
Week 4
Week 5
Week 6
Week 7
Week 8
0 70.0
0 64.3
0 35.3
0 33.3
0 35.7
0 50.0
0 14.3
0 40.0
A.I. Edelstein et al. / The Journal of Arthroplasty xxx (2017) 1e5 Table 5 Correlation Coefficients of Preoperative and Intraoperative Variables With Postoperative Serum Antibiotic Concentrations.
Age Male Diabetes BMI SCr BUN Doses cement Discharged on IV vancomycin
Vancomycin Trough*10
Tobramycin þ Gentamicin Trough*10
0.016 0.062 6.731 0.138 5.939 0.832 3.712 6.238
0.011 0.202 0.340 0.006 0.118 0.004 0.031 0.086
(0.106 to 0.139) (3.177 to 3.300) (0.923-12.540)* (0.032 to 0.308) (15.482 to 3.605) (0.447-1.216)*** (0.759-6.664)* (2.725-9.752)***
(0.019 (0.001 (0.023 (0.017 (0.465 (0.018 (0.160 (0.126
to to to to to to to to
0.003)** 0.405)* 0.703) 0.005) 0.701) 0.027) 0.223) 0.298)
*** P < .001; **P < .01; *P < .05. BMI, body mass index; BUN, blood urea nitrogen; IV, intravenous; SCr, serum creatinine.
vancomycin throughout the study period. We attempted to avoid confounding by controlling for this variable in our analyses. Finally, other studies of systemic absorption from antibiotic cement have found that levels peaked within the first 24 hours after placement. Our intent was to study systemic levels over the duration of ACS treatment, and as such we did not assay serum antibiotic levels in the immediate post-operative period and our data therefore underestimates peak levels of systemic absorption of antibiotics. In sum, use of high-dose ACSs for the treatment of PJI leads to elevated systemic antibiotic concentrations for at least 8 weeks after placement. Absorption of antibiotics from ACS treatment poses a theoretical risk of systemic toxicity, with an increasing body of evidence demonstrating ACS-related renal toxicity. Patients should be closely monitored for complications related to ACS treatment. Further investigation is needed to identify strategies to minimize the risk of systemic toxicity for this population. References [1] Kurtz SM, Ong KL, Schmier J, Mowat F, Saleh K, Dybvik E, et al. Future clinical and economic impact of revision total hip and knee arthroplasty. J Bone Joint Surg Am 2007;89(Suppl 3):144e51. [2] Cui Q, Mihalko WM, Shields JS, Ries M, Saleh KJ. Antibiotic-impregnated cement spacers for the treatment of infection associated with total hip or knee arthroplasty. J Bone Joint Surg Am 2007;89:871e82. [3] Voleti PB, Baldwin KD, Lee GC. Use of static or articulating spacers for infection following total knee arthroplasty: a systematic literature review. J Bone Joint Surg Am 2013;95:1594e9. [4] Osmon DR, Berbari EF, Berendt AR, Lew D, Zimmerli W, Steckelberg JM, et al. Diagnosis and management of prosthetic joint infection: clinical practice guidelines by the Infectious Diseases Society of America. Clin Infect Dis 2013;56:e1e25. [5] Hanssen AD, Osmon DR, Patel R. Local antibiotic delivery systems: where are we and where are we going? Clin Orthop Relat Res 2005:111e4. [6] Anagnostakos K, Wilmes P, Schmitt E, Kelm J. Elution of gentamicin and vancomycin from polymethylmethacrylate beads and hip spacers in vivo. Acta Orthop 2009;80:193e7. [7] 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;13:331e8. [8] Mutimer J, Gillespie G, Lovering AM, Porteus AJ. Measurements of in vivo intra-articular gentamicin levels from antibiotic loaded articulating spacers in revision total knee replacement. Knee 2009;16:39e41. [9] Sterling GJ, Crawford S, Potter JH, Koerbin G, Crawford R. The pharmacokinetics of Simplex-tobramycin bone cement. J Bone Joint Surg Br 2003;85:646e9.
5
[10] Pritchett JW, Bortel DT. Tobramycin-impregnated cement in total hip replacements. Orthop Rev 1992;21:577e9. [11] Marks KE, Nelson CL, Lautenschlager EP. Antibiotic-impregnated acrylic bone cement. J Bone Joint Surg Am 1976;58:358e64. [12] Wentworth SJ, Masri BA, Duncan CP, Southworth CB. Hip prosthesis of antibiotic-loaded acrylic cement for the treatment of infections following total hip arthroplasty. J Bone Joint Surg Am 2002;84-A(Suppl 2):123e8. [13] Springer BD, Lee GC, Osmon D, Haidukewych GJ, Hanssen AD, Jacofsky DJ. Systemic safety of high-dose antibiotic-loaded cement spacers after resection of an infected total knee arthroplasty. Clin Orthop Relat Res 2004: 47e51. [14] Koo KH, Yang JW, Cho SH, Song HR, Park HB, Ha YC, et al. Impregnation of vancomycin, gentamicin, and cefotaxime in a cement spacer for two-stage cementless reconstruction in infected total hip arthroplasty. J Arthroplasty 2001;16:882e92. [15] Luu A, Syed F, Raman G, Bhalla A, Muldoon E, Hadley S, et al. Two-stage arthroplasty for prosthetic joint infection: a systematic review of acute kidney injury, systemic toxicity and infection control. J Arthroplasty 2013;28: 1490e1498.e1. [16] Menge TJ, Koethe JR, Jenkins CA, Wright PW, Shinar AA, Miller GG, et al. Acute kidney injury after placement of an antibiotic-impregnated cement spacer during revision total knee arthroplasty. J Arthroplasty 2012;27: 1221e1227.e1e2. [17] McCormack JP, Jewesson PJ. A critical reevaluation of the “therapeutic range” of aminoglycosides. Clin Infect Dis 1992;14:320e39. [18] Jeffres MN, Isakow W, Doherty JA, Micek ST, Kollef MH. A retrospective analysis of possible renal toxicity associated with vancomycin in patients with health care-associated methicillin-resistant Staphylococcus aureus pneumonia. Clin Ther 2007;29:1107e15. [19] Cunha BA. Aminoglycosides: current role in antimicrobial therapy. Pharmacotherapy 1988;8:334e50. [20] Goodell JA, Flick AB, Hebert JC, Howe JG. Preparation and release characteristics of tobramycin-impregnated polymethylmethacrylate beads. Am J Hosp Pharm 1986;43:1454e61. [21] Wahlig H, Dingeldein E. Antibiotics and bone cements. Experimental and clinical long-term observations. Acta Orthop Scand 1980;51:49e56. [22] Soto-Hall R, Saenz L, Tavernetti R, Cabaud HE, Cochran TP. Tobramycin in bone cement. An in-depth analysis of wound, serum, and urine concentrations in patients undergoing total hip revision arthroplasty. Clin Orthop Relat Res 1983:60e4. [23] Aeng ES, Shalansky KF, Lau TT, Zalunardo N, Li G, Bowie WR, et al. Acute kidney injury with tobramycin-impregnated bone cement spacers in prosthetic joint infections. Ann Pharmacother 2015;49:1207e13. [24] Kalil GZ, Ernst EJ, Johnson SJ, Johannsson B, Polgreen PM, Bertolatus JA, et al. Systemic exposure to aminoglycosides following knee and hip arthroplasty with aminoglycoside-loaded bone cement implants. Ann Pharmacother 2012;46:929e34. [25] Raveh D, Kopyt M, Hite Y, Rudensky B, Sonnenblick M, Yinnon AM. Risk factors for nephrotoxicity in elderly patients receiving once-daily aminoglycosides. QJM 2002;95:291e7. [26] Dahlgren JG, Anderson ET, Hewitt WL. Gentamicin blood levels: a guide to nephrotoxicity. Antimicrob Agents Chemother 1975;8:58e62. [27] Paterson DL, Robson JM, Wagener MM. Risk factors for toxicity in elderly patients given aminoglycosides once daily. J Gen Intern Med 1998;13:735e9. [28] Rybak MJ, Albrecht LM, Boike SC, Chandrasekar PH. Nephrotoxicity of vancomycin, alone and with an aminoglycoside. J Antimicrob Chemother 1990;25:679e87. [29] Bertino Jr JS, Booker LA, Franck PA, Jenkins PL, Franck KR, Nafziger AN. Incidence of and significant risk factors for aminoglycoside-associated nephrotoxicity in patients dosed by using individualized pharmacokinetic monitoring. J Infect Dis 1993;167:173e9. [30] McLaren RL, McLaren AC, Vernon BL. Generic tobramycin elutes from bone cement faster than proprietary tobramycin. Clin Orthop Relat Res 2008;466: 1372e6. [31] Gonzalez Della Valle A, Bostrom M, Brause B, Harney C, Salvati EA. Effective bactericidal activity of tobramycin and vancomycin eluted from acrylic bone cement. Acta Orthop Scand 2001;72:237e40. [32] Penner MJ, Masri BA, Duncan CP. Elution characteristics of vancomycin and tobramycin combined in acrylic bone-cement. J Arthroplasty 1996;11: 939e44. [33] Wahlig H, Dingeldein E, Buchholz HW, Buchholz M, Bachmann F. Pharmacokinetic study of gentamicin-loaded cement in total hip replacements. Comparative effects of varying dosage. J Bone Joint Surg Br 1984;66: 175e9.