- Email: [email protected]
Effect of Antibiotic-Impregnated Bone Cement in Primary Total Knee Arthroplasty
The Journal of Arthroplasty 34 (2019) 2091e2095
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Complications - Infection
Effect of Antibiotic-Impregnated Bone Cement in Primary Total Knee Arthroplasty Hiba K. Anis, MD a, Nipun Sodhi, MD a, Mhamad Faour, MD b, Alison K. Klika, MS a, *, Michael A. Mont, MD a, Wael K. Barsoum, MD b, Carlos A. Higuera, MD b, Robert M. Molloy, MD a a b
Department of Orthopaedic Surgery, Cleveland Clinic Foundation, Cleveland, OH Department of Orthopaedic Surgery, Cleveland Clinic Florida, Weston, FL
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 February 2019 Received in revised form 1 April 2019 Accepted 17 April 2019 Available online 22 April 2019
Background: The purpose of this study is to evaluate the effect of commercially available antibioticimpregnated bone cement (AIBC) on (1) prosthetic joint infections (PJIs) and (2) surgical site infections (SSIs) after primary total knee arthroplasty (TKA). Methods: A review of primary TKAs between 2014 and 2017 from an institutional database was conducted. This identified 12,541 cases which were separated into AIBC (n ¼ 4337) and non-AIBC (8,164) cohorts. Medical records were reviewed for PJIs and SSIs (mean 2-year postoperative period). Infection rates between the cohorts were compared with univariate analyses followed by subanalysis of high risk patients (defined as having 2 or more of the following characteristics: >65 years, body mass index >40, or Charlson Comorbidity Index score >3). To control for confounders, multivariate analyses were performed with regression models adjusted for age, gender, body mass index, comorbidities, year, operative times, and lengths of stay. Results: On univariate analysis, PJI rates were higher in the AIBC cohort (1.0%) compared to the non-AIBC cohort (0.5%, P < .001). Subanalysis of the high risk patients also showed that PJI rates were higher in the AIBC cohort (1.9% vs 0.6%, P < .01). After adjusting for potential confounders, no significant associations between PJIs and AIBC use were found (odds ratio 1.4, 95% confidence interval 0.9-2.3, P ¼ .133). Similarly, no significant differences in SSI rates were observed between the AIBC (2.9%) and non-AIBC cohorts (2.4%, P ¼ .060) and no significant associations between SSIs and AIBC were found with multivariate analysis (odds ratio 1.0, 95% confidence interval CI 0.8-1.3, P ¼ .948). Conclusion: This study found that there was no clinically or statistically significant decrease in infection rates with AIBC in primary TKAs. © 2019 Elsevier Inc. All rights reserved.
Keywords: total knee arthroplasty prosthetic joint infection surgical site infection antibiotic cement postoperative infection
Postoperative infections after total joint arthroplasty (TJA) pose significant economic, social, and psychological burdens on patients, physicians, and healthcare payers. Prosthetic joint infections (PJIs) and surgical site infections (SSIs) can lead to earlier implant failures requiring revision procedures, which in themselves are major risk factors for postoperative complications [1,2]. Therefore, a range of
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.2019.04.033. * Reprint requests: Alison K. Klika, MS, Department of Orthopaedic Surgery, Cleveland Clinic Foundation, 9500 Euclid Avenue/A41, Cleveland, OH 44195. https://doi.org/10.1016/j.arth.2019.04.033 0883-5403/© 2019 Elsevier Inc. All rights reserved.
infection prevention strategies have been implemented over the past few decades [3e6]. Although results of these strategies have generally been favorable, questions remain regarding the standardization of practice as some measures have demonstrated varying levels of success. Antibiotic-impregnated bone cement (AIBC) is one such measure that is often utilized to both treat and prevent infections. The Food and Drug Administration approved low-dose AIBC for use in revision arthroplasty; however, it is frequently used off-label in primary procedures [7e10]. Previous studies using European joint registries had shown that AIBC was superior to systemic antibiotics alone in primary arthroplasty [11e13]. Jamsen et al [13] studied 40,135 primary total knee arthroplasties (TKAs) from the Finnish joint registry between 1997 and 2004 and found a higher
2092
H.K. Anis et al. / The Journal of Arthroplasty 34 (2019) 2091e2095
risk of reoperation for infection (adjusted hazards ratio 1.35, 95% confidence interval [CI] 1.01-1.81) without AIBC. Of note, the patient factors accounted for in the analyses were limited to age, gender, and index diagnoses. However, there is also substantial literature to contradict its routine use in primary TKA [14e17]. A meta-analysis including 3903 primary TKAs from 9 randomized controlled trials found that there were no statistical differences in deep infection rates with AIBC compared to non-AIBC (1.2% vs 1.8%, P ¼ .113) [15]. The debate continues to grow considering the emergence of antimicrobial-resistant bacteria [18e20], compromised cement mechanical properties [21e23], and the additional costs attributed to the widespread use of AIBC [9,24,25]. Thus, current evidence on the efficacy of AIBC in primary TKA is inconclusive [9,26,27]. At the 2018 International Consensus Meeting on Musculoskeletal Infection, no consensus among members was reached regarding the use of AIBC in primary TKA to reduce PJI or SSI risk with 38% agreeing and 58% disagreeing that there is sufficient evidence to support its use [27]. Moreover, with evolving standards for infection control, historical data are becoming less relevant in addressing the efficacy of AIBC for contemporary TKA. Therefore, the purpose of this study is to evaluate the effect of commercially available AIBC on the incidences of (1) PJIs and (2) SSIs after primary TKA. Methods Study Design After Institution Review Board approval, a retrospective review was conducted of patients aged 18-99 years who underwent primary TKAs between January 1, 2014 and December 31, 2017 using an institutional database. A total of 12,674 procedures from 16 centers within our institution were identified. Of these, 133 cases (1%) were excluded as cement was not used. Thus, 12,541 primary TKAs (99%) were included in the final analysis. The mean postoperative follow-up period was 2 ± 1 years (range 0.5-4.5). Data collection entailed electronic queries followed by manual reviews of electronic medical records. Patients were separated into cohorts according to cement type: 8164 in the non-AIBC (65%) group and 4377 (35%) in the AIBC group. The AIBC used at our institution were the commercially available premixed cement packs of 1 g of tobramycin or 1 g of gentamicin in 40 g of polymethyl-methacrylate bone cement. Among the cases in which AIBC was used, tobramycin was used in 2921 (67%) and gentamicin was used in 1451 (33%). Cement was typically mixed using a commercial mixing bowl after implant trialing was performed and applied to the femoral, tibial, and patella components. Surgeries were typically performed using a tourniquet and a standard medial parapatellar approach.
Study Variables A range of variables representing several clinical and patientrelated factors were collected for inclusion in the analysis. Cement type (AIBC or non-AIBC) was recorded for each case as well as the year of surgery, operative times, and lengths of stay (LOS). Operative times were included in the analyses, as they have previously been associated with adverse events including infections [28], and were measured as the time from incision to dressing application. Data on the following patient characteristics were also extracted: age, gender, body mass index (BMI), and existing comorbidities. In order to quantify the extent of patient comorbidities, the Charlson Comorbidity Index (CCI) score was utilized. The CCI score is a measure of 1-year mortality risk according to pre-existing conditions and has previously been linked with TKA outcomes [29e31]. Diagnoses of the following 17 comorbid conditions and their weighted point values were used to calculate CCI scores for each patient: myocardial infarction (1); congestive heart failure (1); peripheral vascular disease (1); connective tissue disease (1); diabetes (1); diabetes with end-organ damage (2); renal disease (2); malignancy; 2); metastatic solid tumor (6); HIV/AIDS (6); cerebrovascular disease (1); dementia (1); hemiplegia (2); pulmonary disease (1); mild liver disease (2); severe liver disease (3); and peptic ulcer disease (1). Additionally, patients were identified as high risk according to age, BMI, and comorbidities. Patients with CCI scores >3 have previously been shown to be at an increased risk for infections and adverse outcomes in TJA [32e34]. Similarly, age >65 years and a BMI >40 have also been associated with increased rates of infections and complications [35e40]. Therefore, patients with 2 or more of the following characteristics were deemed high risk: >65 years, BMI >40, and/or CCI score >3. Data Analysis Baseline characteristics of the cohorts were compared with independent sample t-tests and Pearson’s chi-squared tests. PJI rates between the cohorts were compared with Pearson’s chi-squared tests. Subanalysis including only the high risk patients was then performed to compare PJI rates with AIBC and non-AIBC within this population. Additionally, in order to control for potential confounders, multivariate analysis was performed with binomial logistic regression models. Regression models were adjusted for patient age, gender, BMI, and CCI scores as well as year of surgery, operative times, and LOS. The above analyses were repeated for SSI incidence. All data analyses were performed with SPSS for Windows version 22.0 (IBM Corporation, Armonk, NY). The threshold for statistical significance was maintained at a P-value of less than .05.
Measured Outcomes Results The outcomes measured were PJI incidence and SSI incidence. Patients’ medical records were reviewed by members of the clinical research group for clinician diagnoses of postoperative infections which were classified as PJIs or SSIs. Deep joint infections within 1 year postoperatively with positive cultures and/ or requiring reoperations were classified as PJIs. Infected tissues and synovial fluid were collected in the operating room from the patients who developed infections. Specimens were tested for aerobic and anaerobic organisms and incubated in culture media for a minimum of 7 days. Skin or superficial wound infections treated conservatively with antibiotics alone were classified as SSIs.
Study Population The overall PJI rate was 0.7% (n ¼ 82) and the overall SSI rate was 2.6% (n ¼ 324). There was no association between gender and AIBC use (P ¼ .838). Patients in the AIBC cohort had higher BMIs (P < .001) and CCI scores (P < .001). Mean operative times were longer in the AIBC cohort (116 ± 39 min) compared to the non-AIBC cohort (104 ± 32 min, P < .001). The AIBC cohort had a significantly larger proportion of high risk patients (22%) compared to the non-AIBC cohort (15%, P < .001) indicating that surgeons were more likely to use AIBC in high risk patients (Table 1).
H.K. Anis et al. / The Journal of Arthroplasty 34 (2019) 2091e2095 Table 1 Demographic and Clinical Characteristics of Study Cohorts.
2093
Table 3 Multivariate Logistic Regression Model for Surgical Site Infections.
Characteristic
Non-AIBC (n ¼ 8164)
AIBC (n ¼ 4377)
P-Value
Age (y) Gender Male Female BMI (kg/m2) CCI score High risk patients Year 2014 2015 2016 2017 Operative time (min) LOS (d) PJI SSI
69 ± 9
68 ± 10
<.001a .838b
3068 (65%) 5096 (65%) 32 ± 6 1.5 ± 2.0 1254 (15%)
1653 (35%) 2724 (35%) 34 ± 7 1.9 ± 2.1 942 (22%)
1909 (63%) 1784 (59%) 1676 (55%) 2795 (82%) 104 ± 32 2.5 ± 1.5 39 (0.5%) 195 (2.4%)
1137 (37%) 1242 (41%) 1371 (45%) 627 (18%) 116 ± 39 2.6 ± 1.6 43 (1.0%) 129 (2.9%)
<.001a <.001a <.001b <.001b
<.001a .007a .001b .060b
Statistics are presented as n (row %) or mean ± SD. Bold values indicates statistical significance at P < 0.05. AIBC, antibiotic-impregnated bone cement; BMI, body mass index; CCI, Charlson Comorbidity Index; LOS, length of stay; PJI, prosthetic joint infection; SSI, surgical site infection. a Independent samples t-test. b Pearson’s chi-squared test.
Factors
Cement type (AIBC) Year Gender (female) Age BMI CCI Operative time LOS
Relative Risk
1.008 1.038 0.966 1.017 1.049 1.134 1.006 1.045
95% Confidence Intervals Lower
Upper
0.796 0.938 0.766 1.004 1.032 1.084 1.003 1.045
1.276 1.150 1.218 1.030 1.067 1.187 1.009 1.112
P-Value
.948 .469 .767 .011 <.001 <.001 <.001 .173
Bold values indicates statistical significance at P < 0.05. AIBC, antibiotic-impregnated bone cement; BMI, body mass index; CCI, Charlson Comorbidity Index; LOS, length of stay.
n ¼ 129) and non-AIBC (4.4%, n ¼ 195) cohorts was observed (P < .05). Moreover, there remained no significant associations between AIBC use and SSI incidence after controlling for all the study covariates on multivariate analysis (OR 1.008, 95% CI 0.796-1.276, P ¼ .948) (Table 3). Discussion
Prosthetic Joint Infections AIBC was not found to reduce PJI incidence. Univariate analysis showed significantly higher PJI rates in the AIBC cohort (1.0%, n ¼ 43) compared to the non-AIBC cohort (0.5%, n ¼ 39; P < .001). Similarly, in high risk patients, PJI rates were higher in the AIBC cohort (1.9%, n ¼ 18) compared to the non-AIBC cohort (0.6%, n ¼ 8; P < .01). However, this association was not found to be statistically significant after controlling for potential confounders. The multivariate model adjusted for all the study covariates (ie, age, gender, BMI, CCI score, year of surgery, operative times, and LOS) showed that AIBC did not have a significant effect on PJI incidence (OR 1.422, 95% CI 0.898-2.250, P ¼ .133) (Table 2). The causative organisms for PJIs in both cohorts are provided in Table S1 in the Appendix.
Surgical Site Infections Similarly, AIBC use did not decrease SSI rates. There were no statistically significant differences in SSI rates between the AIBC and non-AIBC cohorts on univariate analysis. The SSI incidence was 2.9% (n ¼ 129) in the AIBC cohort and 2.4% (n ¼ 195) in the nonAIBC cohort (P > .05). Additionally, among the high risk patients only, no significant difference in SSI rates between the AIBC (4.6%,
Table 2 Multivariate Logistic Regression Model for Prosthetic Joint Infections. Factors
Relative Risk
95% Confidence Intervals Lower
Upper
Cement type (AIBC) Year Gender (female) Age BMI CCI score Operative time LOS
1.422 0.938 0.702 0.990 1.029 1.173 1.010 1.133
0.898 0.765 0.449 0.967 0.998 1.078 1.006 1.045
2.250 1.149 1.097 1.013 1.062 1.276 1.015 1.229
P-Value
.133 .534 .120 .396 .066 <.001 <.001 .002
Bold values indicates statistical significance at P < 0.05. AIBC, antibiotic-impregnated bone cement; BMI, body mass index; CCI, Charlson Comorbidity Index; LOS, length of stay.
These findings add to the ongoing debate over the use of prophylactic AIBC in primary TKA. Historical data on total hip arthroplasty from Scandinavian registries dominate current literature on AIBC efficacy [10e12,41] and there is a need for more robust evidence that is applicable to current practices in primary TKA [27]. In 2003, the Food and Drug Administration approved specific, premixed formulations of AIBC (0.5-1.0 g of gentamicin or tobramycin per 40 g of bone cement) for use in the second stage of a 2-stage revision. However, they are commonly used off-label and are often triple the cost of regular bone cement [9,24,25,42,43]. A recent systematic review found that institutions performing more than 1000 TKAs annually could save between $155,000 and $310,000 every year by switching to non-AIBC [43]. Results from this study showed that there are no reductions in PJI or SSI rates after primary TKA with AIBC, even among high risk patients. Therefore, these findings demonstrate that commercially available AIBC does not reduce postoperative infection rates after primary TKA. Given the equivocal literature on AIBC, in 2017, our institution made the change to no longer use AIBC for primary TKAs as part of standard practice. Based on the results of this study, we have continued to implement these changes. Although there is no official singular policy restricting AIBC use in primary TKAs across the institution, this practice has been strongly suggested and supported during quarterly department meetings. This strategy led to a reduction in AIBC use between 2016 and 2017 at our institution by almost two-thirds and has since reduced further to less than 10% of all primary TKAs. For other practices, a major barrier that may exist is most likely reliance on historical data suggesting the superiority of AIBC, which has not necessarily held true in light of recent surgical and medical advances. The findings from this, as well as other recent studies, can be used for further insight into this debate and help effect policy changes. This study had some limitations. Previously published data were used to define high risk patients by age, BMI, and comorbidities. However, although useful for the subanalyses, this definition may be somewhat limited. Another limitation was that surgeon-specific variation in AIBC use was not directly accounted for. However, several patient factors that tend to influence AIBC use were controlled for with multivariate analyses. Moreover, findings from
2094
H.K. Anis et al. / The Journal of Arthroplasty 34 (2019) 2091e2095
this study demonstrated that surgeons were more likely to use AIBC in high risk patients and analyses of these patients also revealed that AIBC did not reduce infection rates. Additionally, a larger sample size may be required to minimize any chance of a type II error. Larger cohorts that also determine specific drug resistance of organisms could also allow effective comparisons of organism profiles between AIBC and non-AIBC PJIs since the present study is underpowered to determine statistical differences. However, to the authors’ knowledge, this study includes the largest AIBC cohort from institution-specific data. Furthermore, using data from a single institution over a relatively short study period also helped maintain consistency in care pathways and limit the potential variability in standards of care. Findings from this study build on existing evidence. Namba et al [44] conducted a review of primary TKAs using a communitybased registry with an AIBC cohort of 2030 patients and found a deep infection rate of 1.4% with AIBC which was significantly higher than the infection rate without AIBC (0.7%, P ¼ .002). However, when a subgroup of diabetic patients was analyzed separately, there were no significant differences in infection rates. Similarly, in our current study, PJI rates were also found to be higher in the AIBC cohort on univariate analysis and after controlling for several potential confounders, no significant association was found between AIBC and infection rates. Additionally, a review of 2293 patients at a single institution between 2003 and 2012 showed that deep infection rates were 0.4% both with and without AIBC (P ¼ 1.000) [17]. Another study of 3292 primary TKAs showed no significant reduction in PJIs at 30 days, 6 months, and 1 year postoperatively even with risk-stratified usage of AIBC [45]. Moreover, a recent meta-analysis of 9 randomized controlled trials found that AIBC did not have a significant effect on infection rates or on clinical knee scores (Knee Society Score and Hospital for Special Surgery score) [15]. There is contrasting literature to suggest that AIBC reduces infection rates in primary TKA. A study of 340 primary TKAs performed between 1994 and 1998 in Taiwan showed that cefuroximeimpregnated cement reduced PJIs after primary TKA [46]. Of note, this study included a small sample of 285 low-risk patients in a markedly different surgical environment compared to current standards in the United States. A retrospective review of 1250 TKAs performed between 2009 and 2012 also showed a significant reduction in PJIs from 3.3% with non-AIBC to 1.3% with AIBC [25]. However, the infection rate in the non-AIBC group in this study was higher than what is reported in current literature and reflects TKAs performed earlier in the study period (2009-2010) compared to those in the AIBC group (2011-2012). There is comparatively little controversy over the efficacy of AIBC in preventing SSIs. Most studies report no significant differences in SSI rates with AIBC [15,16,47] and are in keeping with findings from this study. Although intra-articular antibiotic concentrations can be 10 times what is tolerated systemically [9], antibiotic is unlikely to elute through to superficial layers of the wound [20]. Nevertheless, several other infection prevention strategies have been widely implemented to reduce SSI incidence including, but not limited to, patient optimization, preoperative skin preparation, and negative-pressure wound therapy [4e6]. Conclusion AIBC is currently only indicated for use in revision TJA in the United States; however, it is frequently used prophylactically in primary procedures. This study found that even after accounting for potential confounders, there was no clinically or statistically significant decrease in PJI or SSI rates with commercially available AIBC in primary TKA. With previous reports on the effect of AIBC on
the prevalence of drug-resistant bacteria strains as well as the substantial increase in costs associated with AIBC use, further study on its routine use in primary TKA is warranted. There remains a need for large-scale randomized controlled trials with direct cost analyses to confirm findings from this study. References [1] Ong KL, Lau E, Suggs J, Kurtz SM, Manley MT. Risk of subsequent revision after primary and revision total joint arthroplasty. Clin Orthop Relat Res 2010;468: 3070e6. https://doi.org/10.1007/s11999-010-1399-0. € K, Ha €kkinen U, Remes V. Hospital [2] Pamilo KJ, Peltola M, Paloneva J, M€ akela volume affects outcome after total knee arthroplasty: a nationwide registry analysis of 80 hospitals and 59,696 replacements. Acta Orthop 2015;86:41e7. https://doi.org/10.3109/17453674.2014.977168. [3] Chauveaux D. Preventing surgical-site infections: measures other than antibiotics. Orthop Traumatol Surg Res 2015;101:S77e83. https://doi.org/ 10.1016/j.otsr.2014.07.028. [4] Rezapoor M, Parvizi J. Prevention of periprosthetic joint infection. J Arthroplasty 2015;30:902e7. https://doi.org/10.1016/j.arth.2015.02.044. [5] Ricciardi BF, Bostrom MP, Lidgren L, Ranstam J, Merollini KMD, W-Dahl A. Prevention of surgical site infection in total joint arthroplasty: an international tertiary care center survey. HSS J 2014;10:45e51. https://doi.org/10.1007/ s11420-013-9369-1. [6] Simons MJ, Amin NH, Scuderi GR. Acute wound complications after total knee arthroplasty: prevention and management. J Am Acad Orthop Surg 2017;25: 547e55. https://doi.org/10.5435/JAAOS-D-15-00402. [7] Fish DN, Hoffman HM, Danziger LH. Antibiotic-impregnated cement use in U.S. hospitals. Am J Hosp Pharm 1992;49:2469e74. [8] Heck D, Rosenberg A, Schink-Ascani M, Garbus S, Kiewitt T. Use of antibioticimpregnated cement during hip and knee arthroplasty in the United States. J Arthroplasty 1995;10:470e5. [9] Jiranek WA, Hanssen AD, Greenwald AS. Antibiotic-loaded bone cement for infection prophylaxis in total joint replacement. J Bone Joint Surg Am 2006;88:2487e500. https://doi.org/10.2106/JBJS.E.01126. [10] Josefsson G, Gudmundsson G, Kolmert L, Wijkstrom S. Prophylaxis with systemic antibiotics versus gentamicin bone cement in total hip arthroplasty. A five-year survey of 1688 hips. Clin Orthop Relat Res 1990:173e8. [11] Engesaeter LB, Lie SA, Espehaug B, Furnes O, Vollset SE, Havelin LI. Antibiotic prophylaxis in total hip arthroplasty: effects of antibiotic prophylaxis systemically and in bone cement on the revision rate of 22,170 primary hip replacements followed 0-14 years in the Norwegian Arthroplasty Register. Acta Orthop Scand 2003;74:644e51. https://doi.org/10.1080/000164 70310018135. [12] Espehaug B, Engesaeter LB, Vollset SE, Havelin LI, Langeland N. Antibiotic prophylaxis in total hip arthroplasty. Review of 10,905 primary cemented total hip replacements reported to the Norwegian arthroplasty register, 1987 to 1995. J Bone Joint Surg Br 1997;79:590e5. [13] Jamsen E, Huhtala H, Puolakka T, Moilanen T. Risk factors for infection after knee arthroplasty. A register-based analysis of 43,149 cases. J Bone Joint Surg Am 2009;91:38e47. https://doi.org/10.2106/JBJS.G.01686. [14] Bohm E, Zhu N, Gu J, de Guia N, Linton C, Anderson T, et al. Does adding antibiotics to cement reduce the need for early revision in total knee arthroplasty? Clin Orthop Relat Res 2014;472:162e8. https://doi.org/10.1007/ s11999-013-3186-1. [15] Kleppel D, Stirton J, Liu J, Ebraheim NA. Antibiotic bone cement’s effect on infection rates in primary and revision total knee arthroplasties. World J Orthop 2017;8:946e55. https://doi.org/10.5312/wjo.v8.i12.946. [16] Schiavone Panni A, Corona K, Giulianelli M, Mazzitelli G, Del Regno C, Vasso M. Antibiotic-loaded bone cement reduces risk of infections in primary total knee arthroplasty? A systematic review. Knee Surg Sports Traumatol Arthrosc 2016;24:3168e74. https://doi.org/10.1007/s00167-016-4301-0. [17] Wang H, Qiu G-X, Lin J, Jin J, Qian W-W, Weng X-S. Antibiotic bone cement cannot reduce deep infection after primary total knee arthroplasty. Orthopedics 2015;38:e462e6. https://doi.org/10.3928/01477447-20150603-52. [18] Hansen EN, Adeli B, Kenyon R, Parvizi J. Routine use of antibiotic laden bone cement for primary total knee arthroplasty: impact on infecting microbial patterns and resistance profiles. J Arthroplasty 2014;29:1123e7. [19] Hope PG, Kristinsson KG, Norman P, Elson RA. Deep infection of cemented total hip arthroplasties caused by coagulase-negative staphylococci. J Bone Joint Surg Br 1989;71:851e5. [20] Wininger DA, Fass RJ. Antibiotic-impregnated cement and beads for orthopedic infections. Antimicrobial Agents Chemother 1996;40:2675e9. [21] Armstrong MS, Spencer RF, Cunningham JL, Gheduzzi S, Miles AW, Learmonth ID. Mechanical characteristics of antibiotic-laden bone cement. Acta Orthop Scand 2002;73:688e90. https://doi.org/10.1080/00016470 2321039697. [22] Dunne NJ, Hill J, McAfee P, Kirkpatrick R, Patrick S, Tunney M. Incorporation of large amounts of gentamicin sulphate into acrylic bone cement: effect on handling and mechanical properties, antibiotic release, and biofilm formation. Proc Inst Mech Eng H 2008;222:355e65. https://doi.org/10.1243/0954 4119JEIM355.
H.K. Anis et al. / The Journal of Arthroplasty 34 (2019) 2091e2095 [23] Klekamp J, Dawson JM, Haas DW, DeBoer D, Christie M. The use of vancomycin and tobramycin in acrylic bone cement: biomechanical effects and elution kinetics for use in joint arthroplasty. J Arthroplasty 1999;14: 339e46. [24] Gutowski CJ, Zmistowski BM, Clyde CT, Parvizi J. The economics of using prophylactic antibiotic-loaded bone cement in total knee replacement. Bone Joint J 2014;96-B:65e9. https://doi.org/10.1302/0301-620X.96B1.31428. [25] Sanz-Ruiz P, Matas-Diez JA, Sanchez-Somolinos M, Villanueva-Martinez M, Vaquero-Martin J. Is the commercial antibiotic-loaded bone cement useful in prophylaxis and cost saving after knee and hip joint arthroplasty? The transatlantic paradox. J Arthroplasty 2017;32:1095e9. https://doi.org/ 10.1016/j.arth.2016.11.012. [26] Baeza J, Cury MB, Fleischman A, Ferrando A, Fuertes M, Goswami K, et al. General assembly, prevention, local antimicrobials: proceedings of international consensus on orthopedic infections. J Arthroplasty 2019;34:S75e84. https://doi.org/10.1016/j.arth.2018.09.056. [27] Fillingham Y, Parsa A, Oshkukov S, Greenward AS. International consensus meeting on musculoskeletal infection. Part II Hip and Knee. Section 1.4. 2018. Philadelphia, Available at: https://icmphilly.com/questions/icm-2018-hipand-knee-document/. [28] Bohl DD, Ondeck NT, Darrith B, Hannon CP, Fillingham YA, Della Valle CJ. Impact of operative time on adverse events following primary total joint arthroplasty. J Arthroplasty 2018;33:2256e2262.e4. https://doi.org/10.1016/ j.arth.2018.02.037. [29] Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373e83. [30] Cher EWL, Tay KS, Zhang K, Tan SB, Howe ST, Koh JSB. The effect of comorbidities and age on functional outcomes after total knee arthroplasty in the octogenarian: a matched cohort study. Geriatr Orthop Surg Rehabil 2018;9. https://doi.org/10.1177/2151459318769508. 2151459318769508. [31] Elmallah RDK, Cherian JJ, Robinson K, Harwin SF, Mont MA. The effect of comorbidities on outcomes following total knee arthroplasty. J Knee Surg 2015;28:411e6. https://doi.org/10.1055/s-0035-1549023. [32] Aynardi M, Jacovides CL, Huang R, Mortazavi SMJ, Parvizi J. Risk factors for early mortality following modern total hip arthroplasty. J Arthroplasty 2013;28:517e20. https://doi.org/10.1016/j.arth.2012.06.040. [33] Rasouli MR, Restrepo C, Maltenfort MG, Purtill JJ, Parvizi J. Risk factors for surgical site infection following total joint arthroplasty. J Bone Joint Surg Am 2014;96:e158. https://doi.org/10.2106/JBJS.M.01363. [34] Tay KS, Cher EWL, Zhang K, Tan SB, Howe ST, Koh JSB. Comorbidities have a greater impact than age alone in the outcomes of octogenarian total knee arthroplasty. J Arthroplasty 2017;32:3373e8. https://doi.org/10.1016/ j.arth.2017.05.041.
2095
[35] SooHoo NF, Li Z, Chan V, Chenok K, Bozic KJ. The importance of risk adjustment in reporting total joint arthroplasty outcomes. J Arthroplasty 2016;31: 590e5. https://doi.org/10.1016/j.arth.2015.09.041. [36] Paxton EW, Inacio MCS, Singh JA, Love R, Bini SA, Namba RS. Are there modifiable risk factors for hospital readmission after total hip arthroplasty in a US healthcare system? Clin Orthop Relat Res 2015;473:3446e55. https:// doi.org/10.1007/s11999-015-4278-x. [37] Jung P, Morris AJ, Zhu M, Roberts SA, Frampton C, Young SW. BMI is a key risk factor for early periprosthetic joint infection following total hip and knee arthroplasty. N Z Med J 2017;130:24e34. [38] Baker P, Petheram T, Jameson S, Reed M, Gregg P, Deehan D. The association between body mass index and the outcomes of total knee arthroplasty. J Bone Joint Surg Am 2012;94:1501e8. https://doi.org/10.2106/JBJS.K.01180. [39] Belmont PJ, Goodman GP, Waterman BR, Bader JO, Schoenfeld AJ. Thirty-day postoperative complications and mortality following total knee arthroplasty. J Bone Joint Surg Am 2014;96:20e6. https://doi.org/10.2106/JBJS.M.00018. [40] Alvi HM, Mednick RE, Krishnan V, Kwasny MJ, Beal MD, Manning DW. The effect of BMI on 30 day outcomes following total joint arthroplasty. J Arthroplasty 2015;30:1113e7. https://doi.org/10.1016/j.arth.2015.01.049. [41] Malchau H, Herberts P, Ahnfelt L. Prognosis of total hip replacement in Sweden. Follow-up of 92,675 operations performed 1978-1990. Acta Orthop Scand 1993;64:497e506. [42] Cummins JS, Tomek IM, Kantor SR, Furnes O, Engesaeter LB, Finlayson SRG. Cost-effectiveness of antibiotic-impregnated bone cement used in primary total hip arthroplasty. J Bone Joint Surg Am 2009;91:634e41. https://doi.org/ 10.2106/JBJS.G.01029. [43] King JD, Hamilton DH, Jacobs CA, Duncan ST. The hidden cost of commercial antibiotic-loaded bone cement: a systematic review of clinical results and cost implications following total knee arthroplasty. J Arthroplasty 2018;33: 3789e92. https://doi.org/10.1016/j.arth.2018.08.009. [44] Namba RS, Chen Y, Paxton EW, Slipchenko T, Fithian DC. Outcomes of routine use of antibiotic-loaded cement in primary total knee arthroplasty. J Arthroplasty 2009;24:44e7. https://doi.org/10.1016/j.arth.2009.05.007. [45] Qadir R, Sidhu S, Ochsner JL, Meyer MS, Chimento GF. Risk stratified usage of antibiotic-loaded bone cement for primary total knee arthroplasty: short term infection outcomes with a standardized cement protocol. J Arthroplasty 2014;29:1622e4. https://doi.org/10.1016/j.arth.2014.02.032. [46] Chiu F-Y, Chen C-M, Lin C-FJ, Lo W-H. Cefuroxime-impregnated cement in primary total knee arthroplasty: a prospective, randomized study of three hundred and forty knees. J Bone Joint Surg Am 2002;84-A:759e62. [47] Wang J, Zhu C, Cheng T, Peng X, Zhang W, Qin H, et al. A systematic review and meta-analysis of antibiotic-impregnated bone cement use in primary total hip or knee arthroplasty. PLoS One 2013;8:e82745. https://doi.org/ 10.1371/journal.pone.0082745.
2095.e1
H.K. Anis et al. / The Journal of Arthroplasty 34 (2019) 2091e2095
Appendix
Table S1 Organisms Cultured From Specimens Collected Intraoperatively From Patients With Prosthetic Joint Infections. Non-AIBC Cohort
Frequency
AIBC Cohort
Frequency
Staphylococcus aureus Methicillin-resistant S aureus Coagulase-negative Staphylococcus Staphylococcus epidermidis Peptostreptococcus sp. Pseudomonas aeruginosa Propionibacterium acnes Enterobacter cloacae Proteus mirabilis Streptococcus sp. Corynebacterium sp. Morganella sp. Klebsiella oxytoca Anaerobic gram-positive cocci Pseudallescheria boydii (fungus) Negative
14 7 7 2 4 1 2 1 2 1 2 2 1 1 1 6
Staphylococcus aureus Methicillin-resistant S aureus Coagulase-negative Staphylococcus Staphylococcus epidermidis Peptostreptococcus sp. Pseudomonas aeruginosa Propionibacterium acnes Enterobacter cloacae Proteus mirabilis Streptococcus sp. Escherichia coli Clostridium perfringens Actinomyces europaeus Enterococcus faecalis Serratia sp. Diphtheroid bacilli Corynebacterium sp. Prevotella intermedia Sphingomonas paucimobilis Negative
11 10 5 2 4 3 1 2 1 3 2 1 2 1 3 1 1 1 1 4
AIBC, antibiotic-impregnated bone cement.
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Complications - Infection
Effect of Antibiotic-Impregnated Bone Cement in Primary Total Knee Arthroplasty Hiba K. Anis, MD a, Nipun Sodhi, MD a, Mhamad Faour, MD b, Alison K. Klika, MS a, *, Michael A. Mont, MD a, Wael K. Barsoum, MD b, Carlos A. Higuera, MD b, Robert M. Molloy, MD a a b
Department of Orthopaedic Surgery, Cleveland Clinic Foundation, Cleveland, OH Department of Orthopaedic Surgery, Cleveland Clinic Florida, Weston, FL
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 February 2019 Received in revised form 1 April 2019 Accepted 17 April 2019 Available online 22 April 2019
Background: The purpose of this study is to evaluate the effect of commercially available antibioticimpregnated bone cement (AIBC) on (1) prosthetic joint infections (PJIs) and (2) surgical site infections (SSIs) after primary total knee arthroplasty (TKA). Methods: A review of primary TKAs between 2014 and 2017 from an institutional database was conducted. This identified 12,541 cases which were separated into AIBC (n ¼ 4337) and non-AIBC (8,164) cohorts. Medical records were reviewed for PJIs and SSIs (mean 2-year postoperative period). Infection rates between the cohorts were compared with univariate analyses followed by subanalysis of high risk patients (defined as having 2 or more of the following characteristics: >65 years, body mass index >40, or Charlson Comorbidity Index score >3). To control for confounders, multivariate analyses were performed with regression models adjusted for age, gender, body mass index, comorbidities, year, operative times, and lengths of stay. Results: On univariate analysis, PJI rates were higher in the AIBC cohort (1.0%) compared to the non-AIBC cohort (0.5%, P < .001). Subanalysis of the high risk patients also showed that PJI rates were higher in the AIBC cohort (1.9% vs 0.6%, P < .01). After adjusting for potential confounders, no significant associations between PJIs and AIBC use were found (odds ratio 1.4, 95% confidence interval 0.9-2.3, P ¼ .133). Similarly, no significant differences in SSI rates were observed between the AIBC (2.9%) and non-AIBC cohorts (2.4%, P ¼ .060) and no significant associations between SSIs and AIBC were found with multivariate analysis (odds ratio 1.0, 95% confidence interval CI 0.8-1.3, P ¼ .948). Conclusion: This study found that there was no clinically or statistically significant decrease in infection rates with AIBC in primary TKAs. © 2019 Elsevier Inc. All rights reserved.
Keywords: total knee arthroplasty prosthetic joint infection surgical site infection antibiotic cement postoperative infection
Postoperative infections after total joint arthroplasty (TJA) pose significant economic, social, and psychological burdens on patients, physicians, and healthcare payers. Prosthetic joint infections (PJIs) and surgical site infections (SSIs) can lead to earlier implant failures requiring revision procedures, which in themselves are major risk factors for postoperative complications [1,2]. Therefore, a range of
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.2019.04.033. * Reprint requests: Alison K. Klika, MS, Department of Orthopaedic Surgery, Cleveland Clinic Foundation, 9500 Euclid Avenue/A41, Cleveland, OH 44195. https://doi.org/10.1016/j.arth.2019.04.033 0883-5403/© 2019 Elsevier Inc. All rights reserved.
infection prevention strategies have been implemented over the past few decades [3e6]. Although results of these strategies have generally been favorable, questions remain regarding the standardization of practice as some measures have demonstrated varying levels of success. Antibiotic-impregnated bone cement (AIBC) is one such measure that is often utilized to both treat and prevent infections. The Food and Drug Administration approved low-dose AIBC for use in revision arthroplasty; however, it is frequently used off-label in primary procedures [7e10]. Previous studies using European joint registries had shown that AIBC was superior to systemic antibiotics alone in primary arthroplasty [11e13]. Jamsen et al [13] studied 40,135 primary total knee arthroplasties (TKAs) from the Finnish joint registry between 1997 and 2004 and found a higher
2092
H.K. Anis et al. / The Journal of Arthroplasty 34 (2019) 2091e2095
risk of reoperation for infection (adjusted hazards ratio 1.35, 95% confidence interval [CI] 1.01-1.81) without AIBC. Of note, the patient factors accounted for in the analyses were limited to age, gender, and index diagnoses. However, there is also substantial literature to contradict its routine use in primary TKA [14e17]. A meta-analysis including 3903 primary TKAs from 9 randomized controlled trials found that there were no statistical differences in deep infection rates with AIBC compared to non-AIBC (1.2% vs 1.8%, P ¼ .113) [15]. The debate continues to grow considering the emergence of antimicrobial-resistant bacteria [18e20], compromised cement mechanical properties [21e23], and the additional costs attributed to the widespread use of AIBC [9,24,25]. Thus, current evidence on the efficacy of AIBC in primary TKA is inconclusive [9,26,27]. At the 2018 International Consensus Meeting on Musculoskeletal Infection, no consensus among members was reached regarding the use of AIBC in primary TKA to reduce PJI or SSI risk with 38% agreeing and 58% disagreeing that there is sufficient evidence to support its use [27]. Moreover, with evolving standards for infection control, historical data are becoming less relevant in addressing the efficacy of AIBC for contemporary TKA. Therefore, the purpose of this study is to evaluate the effect of commercially available AIBC on the incidences of (1) PJIs and (2) SSIs after primary TKA. Methods Study Design After Institution Review Board approval, a retrospective review was conducted of patients aged 18-99 years who underwent primary TKAs between January 1, 2014 and December 31, 2017 using an institutional database. A total of 12,674 procedures from 16 centers within our institution were identified. Of these, 133 cases (1%) were excluded as cement was not used. Thus, 12,541 primary TKAs (99%) were included in the final analysis. The mean postoperative follow-up period was 2 ± 1 years (range 0.5-4.5). Data collection entailed electronic queries followed by manual reviews of electronic medical records. Patients were separated into cohorts according to cement type: 8164 in the non-AIBC (65%) group and 4377 (35%) in the AIBC group. The AIBC used at our institution were the commercially available premixed cement packs of 1 g of tobramycin or 1 g of gentamicin in 40 g of polymethyl-methacrylate bone cement. Among the cases in which AIBC was used, tobramycin was used in 2921 (67%) and gentamicin was used in 1451 (33%). Cement was typically mixed using a commercial mixing bowl after implant trialing was performed and applied to the femoral, tibial, and patella components. Surgeries were typically performed using a tourniquet and a standard medial parapatellar approach.
Study Variables A range of variables representing several clinical and patientrelated factors were collected for inclusion in the analysis. Cement type (AIBC or non-AIBC) was recorded for each case as well as the year of surgery, operative times, and lengths of stay (LOS). Operative times were included in the analyses, as they have previously been associated with adverse events including infections [28], and were measured as the time from incision to dressing application. Data on the following patient characteristics were also extracted: age, gender, body mass index (BMI), and existing comorbidities. In order to quantify the extent of patient comorbidities, the Charlson Comorbidity Index (CCI) score was utilized. The CCI score is a measure of 1-year mortality risk according to pre-existing conditions and has previously been linked with TKA outcomes [29e31]. Diagnoses of the following 17 comorbid conditions and their weighted point values were used to calculate CCI scores for each patient: myocardial infarction (1); congestive heart failure (1); peripheral vascular disease (1); connective tissue disease (1); diabetes (1); diabetes with end-organ damage (2); renal disease (2); malignancy; 2); metastatic solid tumor (6); HIV/AIDS (6); cerebrovascular disease (1); dementia (1); hemiplegia (2); pulmonary disease (1); mild liver disease (2); severe liver disease (3); and peptic ulcer disease (1). Additionally, patients were identified as high risk according to age, BMI, and comorbidities. Patients with CCI scores >3 have previously been shown to be at an increased risk for infections and adverse outcomes in TJA [32e34]. Similarly, age >65 years and a BMI >40 have also been associated with increased rates of infections and complications [35e40]. Therefore, patients with 2 or more of the following characteristics were deemed high risk: >65 years, BMI >40, and/or CCI score >3. Data Analysis Baseline characteristics of the cohorts were compared with independent sample t-tests and Pearson’s chi-squared tests. PJI rates between the cohorts were compared with Pearson’s chi-squared tests. Subanalysis including only the high risk patients was then performed to compare PJI rates with AIBC and non-AIBC within this population. Additionally, in order to control for potential confounders, multivariate analysis was performed with binomial logistic regression models. Regression models were adjusted for patient age, gender, BMI, and CCI scores as well as year of surgery, operative times, and LOS. The above analyses were repeated for SSI incidence. All data analyses were performed with SPSS for Windows version 22.0 (IBM Corporation, Armonk, NY). The threshold for statistical significance was maintained at a P-value of less than .05.
Measured Outcomes Results The outcomes measured were PJI incidence and SSI incidence. Patients’ medical records were reviewed by members of the clinical research group for clinician diagnoses of postoperative infections which were classified as PJIs or SSIs. Deep joint infections within 1 year postoperatively with positive cultures and/ or requiring reoperations were classified as PJIs. Infected tissues and synovial fluid were collected in the operating room from the patients who developed infections. Specimens were tested for aerobic and anaerobic organisms and incubated in culture media for a minimum of 7 days. Skin or superficial wound infections treated conservatively with antibiotics alone were classified as SSIs.
Study Population The overall PJI rate was 0.7% (n ¼ 82) and the overall SSI rate was 2.6% (n ¼ 324). There was no association between gender and AIBC use (P ¼ .838). Patients in the AIBC cohort had higher BMIs (P < .001) and CCI scores (P < .001). Mean operative times were longer in the AIBC cohort (116 ± 39 min) compared to the non-AIBC cohort (104 ± 32 min, P < .001). The AIBC cohort had a significantly larger proportion of high risk patients (22%) compared to the non-AIBC cohort (15%, P < .001) indicating that surgeons were more likely to use AIBC in high risk patients (Table 1).
H.K. Anis et al. / The Journal of Arthroplasty 34 (2019) 2091e2095 Table 1 Demographic and Clinical Characteristics of Study Cohorts.
2093
Table 3 Multivariate Logistic Regression Model for Surgical Site Infections.
Characteristic
Non-AIBC (n ¼ 8164)
AIBC (n ¼ 4377)
P-Value
Age (y) Gender Male Female BMI (kg/m2) CCI score High risk patients Year 2014 2015 2016 2017 Operative time (min) LOS (d) PJI SSI
69 ± 9
68 ± 10
<.001a .838b
3068 (65%) 5096 (65%) 32 ± 6 1.5 ± 2.0 1254 (15%)
1653 (35%) 2724 (35%) 34 ± 7 1.9 ± 2.1 942 (22%)
1909 (63%) 1784 (59%) 1676 (55%) 2795 (82%) 104 ± 32 2.5 ± 1.5 39 (0.5%) 195 (2.4%)
1137 (37%) 1242 (41%) 1371 (45%) 627 (18%) 116 ± 39 2.6 ± 1.6 43 (1.0%) 129 (2.9%)
<.001a <.001a <.001b <.001b
<.001a .007a .001b .060b
Statistics are presented as n (row %) or mean ± SD. Bold values indicates statistical significance at P < 0.05. AIBC, antibiotic-impregnated bone cement; BMI, body mass index; CCI, Charlson Comorbidity Index; LOS, length of stay; PJI, prosthetic joint infection; SSI, surgical site infection. a Independent samples t-test. b Pearson’s chi-squared test.
Factors
Cement type (AIBC) Year Gender (female) Age BMI CCI Operative time LOS
Relative Risk
1.008 1.038 0.966 1.017 1.049 1.134 1.006 1.045
95% Confidence Intervals Lower
Upper
0.796 0.938 0.766 1.004 1.032 1.084 1.003 1.045
1.276 1.150 1.218 1.030 1.067 1.187 1.009 1.112
P-Value
.948 .469 .767 .011 <.001 <.001 <.001 .173
Bold values indicates statistical significance at P < 0.05. AIBC, antibiotic-impregnated bone cement; BMI, body mass index; CCI, Charlson Comorbidity Index; LOS, length of stay.
n ¼ 129) and non-AIBC (4.4%, n ¼ 195) cohorts was observed (P < .05). Moreover, there remained no significant associations between AIBC use and SSI incidence after controlling for all the study covariates on multivariate analysis (OR 1.008, 95% CI 0.796-1.276, P ¼ .948) (Table 3). Discussion
Prosthetic Joint Infections AIBC was not found to reduce PJI incidence. Univariate analysis showed significantly higher PJI rates in the AIBC cohort (1.0%, n ¼ 43) compared to the non-AIBC cohort (0.5%, n ¼ 39; P < .001). Similarly, in high risk patients, PJI rates were higher in the AIBC cohort (1.9%, n ¼ 18) compared to the non-AIBC cohort (0.6%, n ¼ 8; P < .01). However, this association was not found to be statistically significant after controlling for potential confounders. The multivariate model adjusted for all the study covariates (ie, age, gender, BMI, CCI score, year of surgery, operative times, and LOS) showed that AIBC did not have a significant effect on PJI incidence (OR 1.422, 95% CI 0.898-2.250, P ¼ .133) (Table 2). The causative organisms for PJIs in both cohorts are provided in Table S1 in the Appendix.
Surgical Site Infections Similarly, AIBC use did not decrease SSI rates. There were no statistically significant differences in SSI rates between the AIBC and non-AIBC cohorts on univariate analysis. The SSI incidence was 2.9% (n ¼ 129) in the AIBC cohort and 2.4% (n ¼ 195) in the nonAIBC cohort (P > .05). Additionally, among the high risk patients only, no significant difference in SSI rates between the AIBC (4.6%,
Table 2 Multivariate Logistic Regression Model for Prosthetic Joint Infections. Factors
Relative Risk
95% Confidence Intervals Lower
Upper
Cement type (AIBC) Year Gender (female) Age BMI CCI score Operative time LOS
1.422 0.938 0.702 0.990 1.029 1.173 1.010 1.133
0.898 0.765 0.449 0.967 0.998 1.078 1.006 1.045
2.250 1.149 1.097 1.013 1.062 1.276 1.015 1.229
P-Value
.133 .534 .120 .396 .066 <.001 <.001 .002
Bold values indicates statistical significance at P < 0.05. AIBC, antibiotic-impregnated bone cement; BMI, body mass index; CCI, Charlson Comorbidity Index; LOS, length of stay.
These findings add to the ongoing debate over the use of prophylactic AIBC in primary TKA. Historical data on total hip arthroplasty from Scandinavian registries dominate current literature on AIBC efficacy [10e12,41] and there is a need for more robust evidence that is applicable to current practices in primary TKA [27]. In 2003, the Food and Drug Administration approved specific, premixed formulations of AIBC (0.5-1.0 g of gentamicin or tobramycin per 40 g of bone cement) for use in the second stage of a 2-stage revision. However, they are commonly used off-label and are often triple the cost of regular bone cement [9,24,25,42,43]. A recent systematic review found that institutions performing more than 1000 TKAs annually could save between $155,000 and $310,000 every year by switching to non-AIBC [43]. Results from this study showed that there are no reductions in PJI or SSI rates after primary TKA with AIBC, even among high risk patients. Therefore, these findings demonstrate that commercially available AIBC does not reduce postoperative infection rates after primary TKA. Given the equivocal literature on AIBC, in 2017, our institution made the change to no longer use AIBC for primary TKAs as part of standard practice. Based on the results of this study, we have continued to implement these changes. Although there is no official singular policy restricting AIBC use in primary TKAs across the institution, this practice has been strongly suggested and supported during quarterly department meetings. This strategy led to a reduction in AIBC use between 2016 and 2017 at our institution by almost two-thirds and has since reduced further to less than 10% of all primary TKAs. For other practices, a major barrier that may exist is most likely reliance on historical data suggesting the superiority of AIBC, which has not necessarily held true in light of recent surgical and medical advances. The findings from this, as well as other recent studies, can be used for further insight into this debate and help effect policy changes. This study had some limitations. Previously published data were used to define high risk patients by age, BMI, and comorbidities. However, although useful for the subanalyses, this definition may be somewhat limited. Another limitation was that surgeon-specific variation in AIBC use was not directly accounted for. However, several patient factors that tend to influence AIBC use were controlled for with multivariate analyses. Moreover, findings from
2094
H.K. Anis et al. / The Journal of Arthroplasty 34 (2019) 2091e2095
this study demonstrated that surgeons were more likely to use AIBC in high risk patients and analyses of these patients also revealed that AIBC did not reduce infection rates. Additionally, a larger sample size may be required to minimize any chance of a type II error. Larger cohorts that also determine specific drug resistance of organisms could also allow effective comparisons of organism profiles between AIBC and non-AIBC PJIs since the present study is underpowered to determine statistical differences. However, to the authors’ knowledge, this study includes the largest AIBC cohort from institution-specific data. Furthermore, using data from a single institution over a relatively short study period also helped maintain consistency in care pathways and limit the potential variability in standards of care. Findings from this study build on existing evidence. Namba et al [44] conducted a review of primary TKAs using a communitybased registry with an AIBC cohort of 2030 patients and found a deep infection rate of 1.4% with AIBC which was significantly higher than the infection rate without AIBC (0.7%, P ¼ .002). However, when a subgroup of diabetic patients was analyzed separately, there were no significant differences in infection rates. Similarly, in our current study, PJI rates were also found to be higher in the AIBC cohort on univariate analysis and after controlling for several potential confounders, no significant association was found between AIBC and infection rates. Additionally, a review of 2293 patients at a single institution between 2003 and 2012 showed that deep infection rates were 0.4% both with and without AIBC (P ¼ 1.000) [17]. Another study of 3292 primary TKAs showed no significant reduction in PJIs at 30 days, 6 months, and 1 year postoperatively even with risk-stratified usage of AIBC [45]. Moreover, a recent meta-analysis of 9 randomized controlled trials found that AIBC did not have a significant effect on infection rates or on clinical knee scores (Knee Society Score and Hospital for Special Surgery score) [15]. There is contrasting literature to suggest that AIBC reduces infection rates in primary TKA. A study of 340 primary TKAs performed between 1994 and 1998 in Taiwan showed that cefuroximeimpregnated cement reduced PJIs after primary TKA [46]. Of note, this study included a small sample of 285 low-risk patients in a markedly different surgical environment compared to current standards in the United States. A retrospective review of 1250 TKAs performed between 2009 and 2012 also showed a significant reduction in PJIs from 3.3% with non-AIBC to 1.3% with AIBC [25]. However, the infection rate in the non-AIBC group in this study was higher than what is reported in current literature and reflects TKAs performed earlier in the study period (2009-2010) compared to those in the AIBC group (2011-2012). There is comparatively little controversy over the efficacy of AIBC in preventing SSIs. Most studies report no significant differences in SSI rates with AIBC [15,16,47] and are in keeping with findings from this study. Although intra-articular antibiotic concentrations can be 10 times what is tolerated systemically [9], antibiotic is unlikely to elute through to superficial layers of the wound [20]. Nevertheless, several other infection prevention strategies have been widely implemented to reduce SSI incidence including, but not limited to, patient optimization, preoperative skin preparation, and negative-pressure wound therapy [4e6]. Conclusion AIBC is currently only indicated for use in revision TJA in the United States; however, it is frequently used prophylactically in primary procedures. This study found that even after accounting for potential confounders, there was no clinically or statistically significant decrease in PJI or SSI rates with commercially available AIBC in primary TKA. With previous reports on the effect of AIBC on
the prevalence of drug-resistant bacteria strains as well as the substantial increase in costs associated with AIBC use, further study on its routine use in primary TKA is warranted. There remains a need for large-scale randomized controlled trials with direct cost analyses to confirm findings from this study. References [1] Ong KL, Lau E, Suggs J, Kurtz SM, Manley MT. Risk of subsequent revision after primary and revision total joint arthroplasty. Clin Orthop Relat Res 2010;468: 3070e6. https://doi.org/10.1007/s11999-010-1399-0. € K, Ha €kkinen U, Remes V. Hospital [2] Pamilo KJ, Peltola M, Paloneva J, M€ akela volume affects outcome after total knee arthroplasty: a nationwide registry analysis of 80 hospitals and 59,696 replacements. Acta Orthop 2015;86:41e7. https://doi.org/10.3109/17453674.2014.977168. [3] Chauveaux D. Preventing surgical-site infections: measures other than antibiotics. Orthop Traumatol Surg Res 2015;101:S77e83. https://doi.org/ 10.1016/j.otsr.2014.07.028. [4] Rezapoor M, Parvizi J. Prevention of periprosthetic joint infection. J Arthroplasty 2015;30:902e7. https://doi.org/10.1016/j.arth.2015.02.044. [5] Ricciardi BF, Bostrom MP, Lidgren L, Ranstam J, Merollini KMD, W-Dahl A. Prevention of surgical site infection in total joint arthroplasty: an international tertiary care center survey. HSS J 2014;10:45e51. https://doi.org/10.1007/ s11420-013-9369-1. [6] Simons MJ, Amin NH, Scuderi GR. Acute wound complications after total knee arthroplasty: prevention and management. J Am Acad Orthop Surg 2017;25: 547e55. https://doi.org/10.5435/JAAOS-D-15-00402. [7] Fish DN, Hoffman HM, Danziger LH. Antibiotic-impregnated cement use in U.S. hospitals. Am J Hosp Pharm 1992;49:2469e74. [8] Heck D, Rosenberg A, Schink-Ascani M, Garbus S, Kiewitt T. Use of antibioticimpregnated cement during hip and knee arthroplasty in the United States. J Arthroplasty 1995;10:470e5. [9] Jiranek WA, Hanssen AD, Greenwald AS. Antibiotic-loaded bone cement for infection prophylaxis in total joint replacement. J Bone Joint Surg Am 2006;88:2487e500. https://doi.org/10.2106/JBJS.E.01126. [10] Josefsson G, Gudmundsson G, Kolmert L, Wijkstrom S. Prophylaxis with systemic antibiotics versus gentamicin bone cement in total hip arthroplasty. A five-year survey of 1688 hips. Clin Orthop Relat Res 1990:173e8. [11] Engesaeter LB, Lie SA, Espehaug B, Furnes O, Vollset SE, Havelin LI. Antibiotic prophylaxis in total hip arthroplasty: effects of antibiotic prophylaxis systemically and in bone cement on the revision rate of 22,170 primary hip replacements followed 0-14 years in the Norwegian Arthroplasty Register. Acta Orthop Scand 2003;74:644e51. https://doi.org/10.1080/000164 70310018135. [12] Espehaug B, Engesaeter LB, Vollset SE, Havelin LI, Langeland N. Antibiotic prophylaxis in total hip arthroplasty. Review of 10,905 primary cemented total hip replacements reported to the Norwegian arthroplasty register, 1987 to 1995. J Bone Joint Surg Br 1997;79:590e5. [13] Jamsen E, Huhtala H, Puolakka T, Moilanen T. Risk factors for infection after knee arthroplasty. A register-based analysis of 43,149 cases. J Bone Joint Surg Am 2009;91:38e47. https://doi.org/10.2106/JBJS.G.01686. [14] Bohm E, Zhu N, Gu J, de Guia N, Linton C, Anderson T, et al. Does adding antibiotics to cement reduce the need for early revision in total knee arthroplasty? Clin Orthop Relat Res 2014;472:162e8. https://doi.org/10.1007/ s11999-013-3186-1. [15] Kleppel D, Stirton J, Liu J, Ebraheim NA. Antibiotic bone cement’s effect on infection rates in primary and revision total knee arthroplasties. World J Orthop 2017;8:946e55. https://doi.org/10.5312/wjo.v8.i12.946. [16] Schiavone Panni A, Corona K, Giulianelli M, Mazzitelli G, Del Regno C, Vasso M. Antibiotic-loaded bone cement reduces risk of infections in primary total knee arthroplasty? A systematic review. Knee Surg Sports Traumatol Arthrosc 2016;24:3168e74. https://doi.org/10.1007/s00167-016-4301-0. [17] Wang H, Qiu G-X, Lin J, Jin J, Qian W-W, Weng X-S. Antibiotic bone cement cannot reduce deep infection after primary total knee arthroplasty. Orthopedics 2015;38:e462e6. https://doi.org/10.3928/01477447-20150603-52. [18] Hansen EN, Adeli B, Kenyon R, Parvizi J. Routine use of antibiotic laden bone cement for primary total knee arthroplasty: impact on infecting microbial patterns and resistance profiles. J Arthroplasty 2014;29:1123e7. [19] Hope PG, Kristinsson KG, Norman P, Elson RA. Deep infection of cemented total hip arthroplasties caused by coagulase-negative staphylococci. J Bone Joint Surg Br 1989;71:851e5. [20] Wininger DA, Fass RJ. Antibiotic-impregnated cement and beads for orthopedic infections. Antimicrobial Agents Chemother 1996;40:2675e9. [21] Armstrong MS, Spencer RF, Cunningham JL, Gheduzzi S, Miles AW, Learmonth ID. Mechanical characteristics of antibiotic-laden bone cement. Acta Orthop Scand 2002;73:688e90. https://doi.org/10.1080/00016470 2321039697. [22] Dunne NJ, Hill J, McAfee P, Kirkpatrick R, Patrick S, Tunney M. Incorporation of large amounts of gentamicin sulphate into acrylic bone cement: effect on handling and mechanical properties, antibiotic release, and biofilm formation. Proc Inst Mech Eng H 2008;222:355e65. https://doi.org/10.1243/0954 4119JEIM355.
H.K. Anis et al. / The Journal of Arthroplasty 34 (2019) 2091e2095 [23] Klekamp J, Dawson JM, Haas DW, DeBoer D, Christie M. The use of vancomycin and tobramycin in acrylic bone cement: biomechanical effects and elution kinetics for use in joint arthroplasty. J Arthroplasty 1999;14: 339e46. [24] Gutowski CJ, Zmistowski BM, Clyde CT, Parvizi J. The economics of using prophylactic antibiotic-loaded bone cement in total knee replacement. Bone Joint J 2014;96-B:65e9. https://doi.org/10.1302/0301-620X.96B1.31428. [25] Sanz-Ruiz P, Matas-Diez JA, Sanchez-Somolinos M, Villanueva-Martinez M, Vaquero-Martin J. Is the commercial antibiotic-loaded bone cement useful in prophylaxis and cost saving after knee and hip joint arthroplasty? The transatlantic paradox. J Arthroplasty 2017;32:1095e9. https://doi.org/ 10.1016/j.arth.2016.11.012. [26] Baeza J, Cury MB, Fleischman A, Ferrando A, Fuertes M, Goswami K, et al. General assembly, prevention, local antimicrobials: proceedings of international consensus on orthopedic infections. J Arthroplasty 2019;34:S75e84. https://doi.org/10.1016/j.arth.2018.09.056. [27] Fillingham Y, Parsa A, Oshkukov S, Greenward AS. International consensus meeting on musculoskeletal infection. Part II Hip and Knee. Section 1.4. 2018. Philadelphia, Available at: https://icmphilly.com/questions/icm-2018-hipand-knee-document/. [28] Bohl DD, Ondeck NT, Darrith B, Hannon CP, Fillingham YA, Della Valle CJ. Impact of operative time on adverse events following primary total joint arthroplasty. J Arthroplasty 2018;33:2256e2262.e4. https://doi.org/10.1016/ j.arth.2018.02.037. [29] Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373e83. [30] Cher EWL, Tay KS, Zhang K, Tan SB, Howe ST, Koh JSB. The effect of comorbidities and age on functional outcomes after total knee arthroplasty in the octogenarian: a matched cohort study. Geriatr Orthop Surg Rehabil 2018;9. https://doi.org/10.1177/2151459318769508. 2151459318769508. [31] Elmallah RDK, Cherian JJ, Robinson K, Harwin SF, Mont MA. The effect of comorbidities on outcomes following total knee arthroplasty. J Knee Surg 2015;28:411e6. https://doi.org/10.1055/s-0035-1549023. [32] Aynardi M, Jacovides CL, Huang R, Mortazavi SMJ, Parvizi J. Risk factors for early mortality following modern total hip arthroplasty. J Arthroplasty 2013;28:517e20. https://doi.org/10.1016/j.arth.2012.06.040. [33] Rasouli MR, Restrepo C, Maltenfort MG, Purtill JJ, Parvizi J. Risk factors for surgical site infection following total joint arthroplasty. J Bone Joint Surg Am 2014;96:e158. https://doi.org/10.2106/JBJS.M.01363. [34] Tay KS, Cher EWL, Zhang K, Tan SB, Howe ST, Koh JSB. Comorbidities have a greater impact than age alone in the outcomes of octogenarian total knee arthroplasty. J Arthroplasty 2017;32:3373e8. https://doi.org/10.1016/ j.arth.2017.05.041.
2095
[35] SooHoo NF, Li Z, Chan V, Chenok K, Bozic KJ. The importance of risk adjustment in reporting total joint arthroplasty outcomes. J Arthroplasty 2016;31: 590e5. https://doi.org/10.1016/j.arth.2015.09.041. [36] Paxton EW, Inacio MCS, Singh JA, Love R, Bini SA, Namba RS. Are there modifiable risk factors for hospital readmission after total hip arthroplasty in a US healthcare system? Clin Orthop Relat Res 2015;473:3446e55. https:// doi.org/10.1007/s11999-015-4278-x. [37] Jung P, Morris AJ, Zhu M, Roberts SA, Frampton C, Young SW. BMI is a key risk factor for early periprosthetic joint infection following total hip and knee arthroplasty. N Z Med J 2017;130:24e34. [38] Baker P, Petheram T, Jameson S, Reed M, Gregg P, Deehan D. The association between body mass index and the outcomes of total knee arthroplasty. J Bone Joint Surg Am 2012;94:1501e8. https://doi.org/10.2106/JBJS.K.01180. [39] Belmont PJ, Goodman GP, Waterman BR, Bader JO, Schoenfeld AJ. Thirty-day postoperative complications and mortality following total knee arthroplasty. J Bone Joint Surg Am 2014;96:20e6. https://doi.org/10.2106/JBJS.M.00018. [40] Alvi HM, Mednick RE, Krishnan V, Kwasny MJ, Beal MD, Manning DW. The effect of BMI on 30 day outcomes following total joint arthroplasty. J Arthroplasty 2015;30:1113e7. https://doi.org/10.1016/j.arth.2015.01.049. [41] Malchau H, Herberts P, Ahnfelt L. Prognosis of total hip replacement in Sweden. Follow-up of 92,675 operations performed 1978-1990. Acta Orthop Scand 1993;64:497e506. [42] Cummins JS, Tomek IM, Kantor SR, Furnes O, Engesaeter LB, Finlayson SRG. Cost-effectiveness of antibiotic-impregnated bone cement used in primary total hip arthroplasty. J Bone Joint Surg Am 2009;91:634e41. https://doi.org/ 10.2106/JBJS.G.01029. [43] King JD, Hamilton DH, Jacobs CA, Duncan ST. The hidden cost of commercial antibiotic-loaded bone cement: a systematic review of clinical results and cost implications following total knee arthroplasty. J Arthroplasty 2018;33: 3789e92. https://doi.org/10.1016/j.arth.2018.08.009. [44] Namba RS, Chen Y, Paxton EW, Slipchenko T, Fithian DC. Outcomes of routine use of antibiotic-loaded cement in primary total knee arthroplasty. J Arthroplasty 2009;24:44e7. https://doi.org/10.1016/j.arth.2009.05.007. [45] Qadir R, Sidhu S, Ochsner JL, Meyer MS, Chimento GF. Risk stratified usage of antibiotic-loaded bone cement for primary total knee arthroplasty: short term infection outcomes with a standardized cement protocol. J Arthroplasty 2014;29:1622e4. https://doi.org/10.1016/j.arth.2014.02.032. [46] Chiu F-Y, Chen C-M, Lin C-FJ, Lo W-H. Cefuroxime-impregnated cement in primary total knee arthroplasty: a prospective, randomized study of three hundred and forty knees. J Bone Joint Surg Am 2002;84-A:759e62. [47] Wang J, Zhu C, Cheng T, Peng X, Zhang W, Qin H, et al. A systematic review and meta-analysis of antibiotic-impregnated bone cement use in primary total hip or knee arthroplasty. PLoS One 2013;8:e82745. https://doi.org/ 10.1371/journal.pone.0082745.
2095.e1
H.K. Anis et al. / The Journal of Arthroplasty 34 (2019) 2091e2095
Appendix
Table S1 Organisms Cultured From Specimens Collected Intraoperatively From Patients With Prosthetic Joint Infections. Non-AIBC Cohort
Frequency
AIBC Cohort
Frequency
Staphylococcus aureus Methicillin-resistant S aureus Coagulase-negative Staphylococcus Staphylococcus epidermidis Peptostreptococcus sp. Pseudomonas aeruginosa Propionibacterium acnes Enterobacter cloacae Proteus mirabilis Streptococcus sp. Corynebacterium sp. Morganella sp. Klebsiella oxytoca Anaerobic gram-positive cocci Pseudallescheria boydii (fungus) Negative
14 7 7 2 4 1 2 1 2 1 2 2 1 1 1 6
Staphylococcus aureus Methicillin-resistant S aureus Coagulase-negative Staphylococcus Staphylococcus epidermidis Peptostreptococcus sp. Pseudomonas aeruginosa Propionibacterium acnes Enterobacter cloacae Proteus mirabilis Streptococcus sp. Escherichia coli Clostridium perfringens Actinomyces europaeus Enterococcus faecalis Serratia sp. Diphtheroid bacilli Corynebacterium sp. Prevotella intermedia Sphingomonas paucimobilis Negative
11 10 5 2 4 3 1 2 1 3 2 1 2 1 3 1 1 1 1 4
AIBC, antibiotic-impregnated bone cement.