Higher Tissue Concentrations of Vancomycin Achieved with Intraosseous Regional Prophylaxis in Revision TKA: a Randomized Controlled Trial

Clin Orthop Relat Res (2018) 476:66-74 DOI 10.1097/JOT.

The John N. Insall Award: Higher Tissue Concentrations of Vancomycin Achieved With Intraosseous Regional Prophylaxis in Revision TKA: A Randomized Controlled Trial

Simon W. Young FRACS, Mei Zhang PhD, Grant A. Moore BSc, Rocco P. Pitto PhD, Henry D. Clarke MD, Mark J. Spangehl MD

Abstract Background In primary TKA, pro- during a prolonged procedure may difference in early postoperative (< 6 phylaxis with low-dose vancomycin lower tissue concentrations. weeks) complications between IORA through intraosseous regional admin- Questions/purposes (1) Does low-dose and systemic IV administration in re- istration (IORA) achieves tissue con- IORA reliably provide equal or higher vision TKA? centrations six to 10 times higher than tissue concentrations of vancomycin Methods Twenty patients undergoing systemic administration and was compared with systemic IV adminis- aseptic revision TKA were randomized shown to provide more effective pro- tration in revision TKA? (2) Are tissue to two groups. The IV group received phylaxis in an animal model. However, concentrations of vancomycin after 1 g systemic IV prophylactic vanco- in revision TKA, the presence of a tib- IORA maintained for the duration of mycin. The IORA group received 500 ial implant may compromise IORA the revision TKA despite a period of mg vancomycin as a bolus injection into injection, and tourniquet deﬂation tourniquet deﬂation? (3) Is there any a tibial intraosseous cannula below an

The institution of one or more of the authors (HDC, MJS) received funding from Vidacare Corp (San Antonio, TX, USA), who manufactured the intraosseous needles used in this study. Clinical Orthopaedics and Related Research® neither advocates nor endorses the use of any treatment, drug, or device. Readers are encouraged to always seek additional information, including FDA-approval status, of any drug or device prior to clinical use. Each author certiﬁes that his or her institution approved the human protocol for this investigation, that all investigations were conducted in conformity with ethical principles of research, and that informed consent for participation in the study was obtained. Procedures and sample collection performed at Mayo Clinic, AZ, USA. Sample analysis performed at Canterbury Health Laboratories, Christchurch, New Zealand. Clinical Trial Registration: NCT02020031.

S. W. Young, R. P. Pitto Department of Orthopaedics, North Shore Hospital, Auckland, New Zealand; and the Department of Surgery, University of Auckland, Auckland, New Zealand

M. Zhang Clinical Pharmacology, Department of Medicine, University of Otago, Christchurch, New Zealand

G. A. Moore Toxicology, Canterbury Health Laboratories, Christchurch, New Zealand

H. D. Clarke, M. J. Spangehl Department of Orthopaedics, Mayo Clinic, Scottsdale, AZ, USA

S. W. Young (✉), Department of Orthropaedics, North Shore Hospital, PrivateBag93-503,Auckland0740NewZealand, email: simon.young@ auckland.ac.nz

All ICMJE Conﬂict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on ﬁle with the publication and can be viewed on request.

Copyright Ó 2017 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited. Volume 476, Number 1 Intraosseous Regional Prophylaxis Revision TKA 67 2017 Knee Society Proceedings

inflated thigh tourniquet before skin in- cohort studies, which might focus on administration [29] and is reliably cision. In all patients receiving IORA, broader safety endpoints in more di- successful in primary TKA [32, 33]. intraosseous tibial injection was techni- verse patient populations. We believe However, in revision TKA, it is unclear cally possible despite the presence of that these studies should evaluate if intraosseous injection into the prox- a tibial implant. Mean procedure length patients undergoing revision TKA in imal tibia will be technically possible was 3.5 hours in both groups. Mean particular, because the risk of infection in the presence of a tibial implant. initial tourniquet inflation was 1.5 hours is greater than in patients undergoing Additionally, revision TKA is often with a second inflation for a mean of 35 primary TKA. prolonged and the tourniquet may be minutes during cementation. During the Level of Evidence Level I, therapeutic deflated during the procedure. This procedure, subcutaneous fat and bone study. removes the circulatory restriction of samples were taken at regular intervals. the antibiotic to the affected limb, po- Tissue vancomycin concentrations were tentially lowering tissue concen- measured using high-performance liq- trations at the surgical site after uid chromatography. deflation. Because the goal of pro- Results Overall geometric mean tissue Introduction phylaxis is to provide adequate con- concentration of vancomycin in fat centrations of antibiotic “from the time samples was 3.7 mg/g (95% confidence eriprosthetic joint infection (PJI) of incision to the time of closure” [2], interval [CI], 2.6-5.2) in the IV group is more common after revision this may decrease the effectiveness of versus 49.3 mg/g in the IORA group PTKA with reported rates as high IORA in revision TKA. (95% CI, 33.2-73.4; ratio between means as 9% [18]. Such PJIs are more chal- This study was performed to com- 13.5; 95% CI, 8.2-22.0; p < 0.001); mean lenging to treat, because revision pare tissue concentrations of vancomy- tissue concentrations in femoral bone implants often involve the use of cin through systemic IV administration were 6.4 mg/g (95% CI, 4.5-9.2) in the stems, cones, and augments, making versus IORA in revision TKA, in which IV group versus 77.1 mg/g (95% CI, thorough debridement´ or removal dif- the risk of PJI is higher. We aimed to 42.4-140) in the IORA group (ratio be- ficult. Prophylactic antibiotics reduce answer the following questions: (1) tween means 12.0; 95% CI, 6.2-23.2; p < the risk of developing PJI [11, 14]; Does low-dose IORA reliably provide 0.001). Vancomycin concentrations in however, bacterial resistance to com- equal or higher tissue concentrations of the final subcutaneous fat sample taken mon prophylactic antibiotics such as vancomycin compared with systemic IV before closure were 5.3 times higher in cephalosporins is increasing [19, 27, administration in revision TKA? (2) Are the IORA group versus the IV group 30]. Vancomycin has been proposed as tissue concentrations of vancomycin (mean 6 SD, 18.2 6 11.6 mg/g IORA an alternative [26]; however, it requires after IORA maintained for the duration versus 3.6 6 2.5 mg/g; p < 0.001). The a prolonged administration time, can of the revision TKA despite a period of intraarticular concentration of vancomy- cause systemic toxicity, and risks pro- tourniquet deflation? (3) Is there any cin on postoperative Day 1 drain samples moting further antibiotic resistance. difference in early postoperative (< 6 was not different between the two groups Low-dose prophylactic vancomycin weeks) complications between IORA with the numbers available (mean through intraosseous regional admin- and systemic IV administration in re- 4.6 mg/L in the IV group versus 6.6 mg/g istration (IORA) may mitigate these vision TKA? in the IORA group; mean difference issues and in primary TKA achieves 2.0 mg/g; 95% CI, 6.2-23.2; p = 0.08). tissue concentrations six to 10 times Conclusions IORA administration of higher than systemic administration vancomycin in patients undergoing [32]. In an animal model of TKA, Patients and Methods revision TKA resulted in tissue con- IORA was also shown to provide more centrations of vancomycin five to 20 effective prophylaxis against PJI [31]. Patients undergoing unilateral revision times higher than systemic IV admin- In TKA prophylaxis, IORA TKA at a single tertiary institution istration despite the lower dose. High involves intraosseous injection of were eligible for enrollment in this tissue concentrations were maintained antibiotics into the proximal tibia after prospective, randomized controlled throughout the procedure despite a pe- tourniquet inflation and before skin trial. Ethical approval was obtained riod of tourniquet deflation. These incision. Even in adults, intraosseous from the institutional ethical review preliminary results justify prospective injection is equivalent to IV board, and the trial and protocol were

Copyright Ó 2017 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited. 68 Young et al. Clinical Orthopaedics and Related Research® 2017 Knee Society Proceedings

registered with ClinicalTrials.gov intraosseous cannula, below an inflated antibiotic prophylaxis regardless of (Identifier: NCT02020031). Inclusion thigh tourniquet, immediately (< 2 randomization. Two patients (one from criteria were patients undergoing minutes) before skin incision. No fur- the IORA group and one from the IV single-stage aseptic revision TKA with ther postoperative doses of vancomy- group) were withdrawn from the study exchange of both tibial and femoral cin were administered in either group, without samples being taken after an components. Exclusion criteria were but cefazolin was continued for 24 intraoperative decision not to proceed previous or current PJI, known hyper- hours. In all patients receiving IORA, with full revision of both components, sensitivity to vancomycin, and major intraosseous tibial injection was tech- leaving 20 patients for analysis (10 in cardiac or respiratory disease. All nically possible despite the presence of the IV group and 10 in the IORA procedures were performed by one of a tibial implant. Although no patients group). Despite randomization, there two fellowship-trained arthroplasty in either group had large cortical per- were more females in the IORA group surgeons (HDC, MJS). forations or massive osteolysis of the (seven versus five IORA) and the mean Between January 2014 and April proximal tibia, in this situation, intra- time between the first and second 2015, 22 patients were enrolled by osseous injection into the medial mal- tourniquet inflations was longer in the a trained research nurse (DLR) in an leolus was the planned alternative IORA group (61 minutes versus 37 outpatient setting. Patients were, using option. Randomization occurred at the minutes IORA). There were no other computer-generated random alloca- time of consent (typically the day be- major differences between groups tions placed in numbered, opaque, fore surgery) to allow the appropriate (Table 1). Intention-to-treat analysis of sealed envelopes, in two groups. The order to be placed to allow the IV complications did not alter results (no IV group received 1 g systemic IV vancomycin to be infused at least 1 complications occurred in the with- prophylactic vancomycin over a 1- hour preoperatively or the IO vanco- drawn patients), and the per-protocol hour infusion into an arm vein, timed mycin to be delivered to the operating analysis is presented. All patients in to finish immediately before tourniquet room for IO administration. Both both groups had estimated glomerular inflation. The IORA group received groups received 2 g systemic cefazolin filtration rates > 50 preoperatively and 500 mg vancomycin in 150 mL saline 15 minutes before tourniquet inflation at the time of discharge. All patients as a bolus injection through a tibial to ensure all patients received effective received a general anesthetic combined

Table 1. Patient demographics Demographic 1 g systemic (n = 10) 500 mg IORA (n = 10) Males 5 3 Females 5 7 Age (years) 68 (54-82) 69 (43-83) BMI (kg/m2) 33 (22-42) 32 (26-42) Indication for revision† (n) Aseptic loosening 5 5 Polyethylene wear 0 2 Instability 4 4 Malrotation/malalignment 4 3 First tourniquet time (minutes)* 94 (85-108) 91 (89-96) Tourniquet deﬂation time (minutes)* 37 (15-88) 61 (18-109) Second tourniquet time (cementation; minutes)* 35 (25-48) 35 (21-52) Total procedure time (minutes skin to skin) 212 (177-282) 219 (167-263) ASA score (range) 2.4 (2-3) 2.7 (2-3) Values given as mean with range in parentheses. *excludes three patients (one systemic, two IORA) in whom a tourniquet was used for 120 minutes and not reinﬂated. †some patients had more than one indication. IORA = Intraosseous Regional Administration; BMI = Body Mass Index; ASA = American Society of Anesthesiologists.

with a periarticular injection of local drain. All patient samples were ana- such as methicillin-resistant Staphylo- anesthetic; eight patients (four in each lyzed in duplicate, and laboratory coccus aureus and coagulase-negative group) also received a peripheral nerve analysis was carried out blinded as to staphylococci [8]. An animal study of block. Revision TKA was performed the randomization group. IORA vancomycin supports this [31]; using components from one of two therefore, the differences used in our suppliers (Stryker Inc, Mahwah, NJ, power analysis are likely to be clinically USA; Zimmer, Warsaw, IN, USA). Power Calculation relevant. Three intraoperative tissue cultures were taken in all patients to ex- Data from a previous randomized trial clude PJI. comparing IORA of 250 mg versus Statistical Analysis The revision TKA was performed systemic administration of 1 g vanco- with the tourniquet initially inflated for mycin [32] showed mean tissue Means, SDs, and the 95% confidence exposure and implant removal, de- concentrations of vancomycin in sub- limits were calculated for the concen- flated, and then reinflated for implant cutaneous fat at different collection trations in the different samples. Dif- cementation. Tobramycin cement was points ranged from 8.1 (SD 5.6) to 19.4 ferent tissue samples were pooled used in all patients in both groups. (SD 11.7) mg/g in the intraosseous according to surgical steps (Appendix Patients were monitored for clinical group and from 2.4 (SD 1.5) to 4.4 (SD 1). Coefficients of variation of con- signs of red man syndrome, particu- 2.0) mg/g in the IV systemic group; centration level were summarized at larly after tourniquet deflation. An an- thus, the concentration of vancomycin each surgical step for comparison be- tihistamine was available for use if was approximately 3.3 times higher in tween the two groups. Repeated- required. Mean procedure length was the IORA group. In bone samples, the measures analysis of variance was 3.5 hours in both groups. The mean difference in vancomycin concentra- used to compare the average level of initial tourniquet inflation was 1.5 tion was 4.5-fold. Using these data, concentration across time between hours with a second inflation for a priori power analysis calculated 10 groups adjusted by body mass index, a mean of 35 minutes during implant patients in each arm would provide age, and length of the surgical pro- cementation. In three patients (one > 90% statistical power to detect the cedure; the Shapiro-Wilk test was used systemic, two IORA), the tourniquet expected fold difference in sub- to assess the normality of the residuals. was used for 120 minutes and not cutaneous fat and bone concentrations Adverse events were tallied for each reinflated. During the procedure, sub- at a 5% significance level if IORA group and reported descriptively. cutaneous fat and bone samples (ap- doses 25% (250 mg) of the systemic proximately 0.5 cm3) were taken at dose (1 g) were used. As a result of the regular intervals until skin closure; longer nature of revision surgery, we a total of six fat and four bone samples chose to use a higher IORA dose of 500 Results were taken for each patient at the same mg; therefore, this power analysis operative steps in both groups (see represents a conservative estimate Higher tissue and bone concentrations Appendix 1, Supplemental Digital likely to overestimate the number of consistently were achieved in patients Content 1). All bone samples were patients required. treated with IORA. The overall geo- taken from the femur, distant from the There are limited data on the phar- metric mean of average tissue con- tibial intraosseous injection site. Tissue macodynamics of vancomycin for sur- centrations of vancomycin in fat samples were stored at -90° C until gical prophylaxis; however, in treatment samples was 3.7 mg/g (95% confidence they were analyzed. models of infection, the pharmacoki- interval [CI], 2.6-5.2) in the IV group Vancomycin concentrations were netic-pharmacodynamic parameter most versus 49.3 mg/g in the IORA group determined by liquid chromatography predictive of efficacy is the area under the (95% CI, 33.2-73.4; ratio between coupled with tandem mass spectrome- concentration time curve divided by means 13.5; 95% CI, 8.2-22.0; p < try using a validated technique that has the minimum inhibitory concentration 0.001; Table 2). Overall geometric been previously described [32, 34]. On (MIC) [7]. Increased vancomycin tissue mean of average tissue concentrations postoperative Day 1, intraarticular concentrations can therefore be expected in femoral bone was 6.4 mg/g (95% CI, vancomycin concentrations were de- to enhance effectiveness, especially in 4.5-9.2) in the IV group versus 77.1 termined from a sample taken from the organisms with MICs of 1 g/L or more mg/g (95% CI, 42.4-140) in the IORA

Copyright Ó 2017 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited. 70 Young et al. Clinical Orthopaedics and Related Research® 2017 Knee Society Proceedings

group (ratio between means 12.0; 95% postoperative Day 1 drain samples was etiology of the foot drop was unknown, CI, 6.2-23.2; p < 0.001; Fig. 1A-B; see not different between the two groups although the patient was obese and sur- Appendix 2, Supplemental Digital with the numbers available (mean 4.6 gical exposure was difficult. This com- Content 2, and Appendix 3, Supple- mg/L in the IV group versus 6.6 mg/g in plication has not previously been mental Digital Content 3). the IORA group; mean difference 2.0 reported with IORA and was not thought Sustained levels of vancomycin mg/g; 95% CI, 6.2-23.2; p = 0.08; Fig. 3). to be related. There were no other reop- remained higher throughout the proce- No patient in either group developed erations in either group. dures in patients treated with IORA. symptoms of red man syndrome. There Vancomycin concentrations in the final were no thromboemobolic complica- subcutaneous fat sample taken before tions or deep or superficial infections in Discussion closure were 5.3 times higher in the either group. One patient in the IORA IORA group versus the IV group (mean, group developed foot drop postopera- PJI is more common in revision than 18.2 6 11.6 mg/g IORA versus 3.6 6 2.5 tively and underwent exploration and primary TKA [18, 25] and has been mg/g; p < 0.001; Fig. 2). The intra- decompression of the common peroneal reported to account for > 44% of rere- articular concentration of vancomycin on nerve 3.5 months postoperatively. The visions after revision TKA [17].

Table 2. Mean tissue concentrations of vancomycin at each sample point 1 g systemic 500 mg IORA Time Mean concentration Time Mean concentration (minutes; 6 SD) (mg/g; 6 SD) (minutes; 6 SD) (mg/g; 6 SD) p value Subcutaneous fat 1 (S1) 2 3.2 3 94 < 0.001 (1) (1.8) (4) (69) Subcutaneous fat 2 (S2) 34 5.0 30 88 < 0.001 (7) (2.9) (6) (131) Subcutaneous fat 3 (S4) 50 4.2 48 69 < 0.001 (10) (2.5) (12) (50) Subcutaneous fat 4 (S6) 115 4.7 119 173 < 0.001 (30) (2.6) (30) (445) Subcutaneous fat 5 (S8) 145 4.0 151 249 < 0.001 (34) (2.2) (36) (639) Subcutaneous fat 6 (S10) 180 3.6 193 18 < 0.001 (64) (2.5) (82) (11.6)

Bone sample 1 (S3) 34 7.9 30 91 < 0.001 (7) (5.7) (6) (77) Bone sample 2 (S5) 50 8.6 48 193 < 0.001 (10) (5.9) (12) (191) Bone sample 3 (S7) 115 5.0 119 60 < 0.001 (30) (2.3) (30) (63) Bone sample 4 (S9) 145 7.1 151 63 < 0.001 (34) (4.4) (36) (62)

Times are given as minutes after surgical incision. diﬀerences in mean tissue concentrations between the two groups were statistically signiﬁcant (p < 0.0001) for all comparison points after adjustment by gender, age, body mass index, American Society of Anesthesiologists score, interaction between groups, and time from incision. IORA = Intraosseous Regional Administration.

A limitation of this study is that the power analysis was based on the tissue concentration of vancomycin and not the development of PJI, because the numbers required to detect a difference in PJI incidence between the two tech- niques would be prohibitive. However, the pharmacokinetic-pharmacodynamic parameter most predictive of efficacy of vancomycin is the area under the concentration time curve divided by the MIC [7]; therefore, higher vancomycin tissue concentrations are likely to enhance effectiveness. A recent animal study supports this, finding low-dose IORA vancomycin prophylaxis to be more effective than standard-dose systemic vancomycin in preventing PJI in a murine model of TKA [31]. Second, we used a standard 1-g dose of vancomycin, whereas some authors have ad- vocated weight-based dosing (eg, 15 mg/L) to ensure adequate tissue concentrations are achieved [3]. However, a 1-g dose for orthopaedic prophylaxis is commonly reported [8, 22, 26]and given the magnitude of the difference seen, the use of a weight-based systemic dose would have been unlikely to alter the findings of this study. Third, randomization envelopes were opened the day before surgery to allow pharmacy preparation of the appropriate vancomycin administration method. To avoid potential subversion of randomization, the envelopes were opened by an in- dependent researcher and surgeons were not informed of the randomization until the day of surgery. Also, although Fig. 1 Intraosseous injection was performed after tourniquet inflation. After injection, the needle is removed and the injection site sample analysis was performed in covered with Ioban (3M, St Paul, MN, USA) before proceeding with the ablindedfashion,thesurgeonwasnot skin incision and surgery. blinded to treatment group potentially biasing interpretation of short-term Measures aiming to reduce PJI there- low-dose vancomycin to be effective in complications. This was not the pri- fore are particularly relevant in re- the setting of revision TKA, reliably mary outcome of the study, however, vision TKA given both the higher providing tissue concentrations five to and there were few complications in incidence and the difficultly in treat- 20 times higher than systemic IV ad- either group. Finally, although we ment in the presence of revision ministration for the duration of the found no difference in complication implants. This study found IORA of procedure. rates between the two groups, the

Copyright Ó 2017 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited. 72 Young et al. Clinical Orthopaedics and Related Research® 2017 Knee Society Proceedings

Fig. 2 A-B (A) Scatterplot showing tissue concentration of vancomycin in subcutaneous fat at various time points after incision. Note the scale is logarithmic. (B) Scatterplot showing tissue concentration of vancomycin in bone at various time points after incision. Note the scale is logarithmic.

number of patients was relatively septic arthritis, again without reported very high vancomycin concentrations small. Two previous studies of IORA complications [23, 24]. seen in the first tissue sample taken in primary TKA have also found no We found intraosseous injection to within minutes of the IORA injection. increase in complications [32, 33], be successful in all patients despite the Regional administration of pro- and in veterinary research, IORA of presence of a tibial implant. Rapid phylactic antibiotics in TKA has been vancomycin and other antibiotics is distribution through the limb circula- previously investigated by de Lalla well described for treatment of equine tion after injection was evident by the et al. [9] comparing IV regional administration (IVRA) of 400 mg teicoplanin given through a foot vein with 800 mg teicoplanin given systemically. They reported tissue concentrations two to 10 times higher in the IVRA group. They evaluated this IVRA protocol in 250 patients undergoing TKA, reporting a 0% PJI rate [10]. In this study, we used the intraosseous route to perform regional administration, and the main advantages of IORA over IVRA are reliability and speed. Can- nulation of a foot vein can be difficult in obese patients and involves expos- ing an area typically covered in sterile drapes. In contrast, intraosseous injection using modern equipment is rapid and reproducible [5], and injec- tions travel directly into the in- travascular space in a manner Fig. 3 Graph showing the intraarticular concentration of vancomycin in drain fluid drawn the morning after surgery. Central line equivalent to IV injection in both represents the median; box represents the 25% and 75% quartiles; adults and children [29]. A small area whiskers represent the range. of cancellous bone is all that is

required, and the presence of tibial the use of a lower vancomycin dose. These preliminary results justify pro- implants in this study did not alter the The lower dose allows bolus adminis- spective cohort studies, which might effectiveness of IORA. In the setting of tration instead of a prolonged systemic focus on broader safety endpoints in severe proximal tibial bone loss, the infusion and minimizes the risk of more diverse patient populations. We IORA technique may not be feasible, systemic complications such as red believe that these studies should eval- although intraosseous injection can man syndrome [20] or nephrotoxicity uate patients undergoing revision TKA also be performed into the distal tibia, [6]. This also avoids the need for pre- in particular, because the risk of in- distal femur, or calcaneus [4, 15, 16]. operative coordination to ensure ap- fection is greater than in patients un- We found higher tissue concen- propriate timing of prophylaxis dergoing primary TKA. trations of vancomycin throughout the administration, because most hospital duration of the procedure in the IORA protocols require vancomycin infu- group, despite intraoperative tourni- sions of 1 g over 1 to 2 hours to avoid Acknowledgments We thank Debra L. Ryan, research assistant, for her help in patient quet release. Revision TKAs are often red man syndrome. recruitment, sample management, and data prolonged, and inflation of the tourni- The very high concentrations seen gathering. quet for the entire procedure risks with IORA raises the question of po- nerve or ischemic injury. Once the tential local toxicity. The in vitro effect References tourniquet is released, vancomycin of high antibiotic concentrations on 1. Antoci V, Adams CS, Hickok NJ, Sha- levels at the operative site can be musculoskeletal cells has been in- piro IM, Parvizi J. Antibiotics for local expected to decrease, although before vestigated in the context of local delivery systems cause skeletal cell tox- this study, the rate at which they did so delivery of antibiotic-impregnated ce- icity in vitro. Clin Orthop Relat Res. was unclear. We found intraarticular ment for the treatment of infection. 2007;462:200–206. 2. Burke JF. The effective period of pre- levels on postoperative Day 1 Antoci et al. reported minimal toxicity ventive antibiotic action in experimental remained above the typical vancomy- to osteoblastic and chondroblastic cell incisions and dermal lesions. Surgery. cin MIC reported for methicillin- lines at vancomycin concentrations of 1961;50:161–168. resistant S aureus (1.0 mg/mL) and 250 mg/mL with a reduction in cellular 3. Catanzano A, Phillips M, Dubrovskaya coagulase-negative staphylococci (2.0 proliferation becoming apparent at Y, Hutzler L, Bosco J. The standard one gram dose of vancomycin is not adequate mg/mL) [28]. There is likely to be concentrations > 2000 mg/mL [1]. prophylaxis for MRSA. Iowa Orthop J. a depot effect of the initial high tissue Similarly, Rathbone et al. reported 2014;34:111–117. concentrations, leading antibiotic to be vancomycin to be the least toxic of 21 4. Clem M, Tierney P. Intraosseous infu- released gradually into the systemic antibiotics tested with no effect on os- sions via the calcaneus. Resuscitation. circulation after tourniquet deflation teoblast survival or metabolic function 2004;62:107–112. 5. Cooper BR, Mahoney PF, Hodgetts TJ, [23]. A potential weakness of the until exposed to concentrations in ex- Mellor A. Intra-osseous access (EZ-IO) IORA technique is that many surgeons cess of 2000 mg/mL for 10 to 14 days for resuscitation: UK military combat routinely continue antibiotics for 24 [21]. The IORA tissue concentrations experience. J R Army Med Corps. 2007; hours postoperatively, and further in our study were well below these 153:314–316. systemic vancomycin doses would levels and the duration of exposure 6. Courtney PM, Melnic CM, Zimmer Z, Anari J, Lee G-C. Addition of vanco- thus still be required after IORA. shorter, suggesting local toxicity is mycin to cefazolin prophylaxis is asso- However, randomized trials have unlikely to occur. ciated with acute kidney injury after shown no difference in infection rates In conclusion, IORA administration primary joint arthroplasty. Clin Orthop between a single preoperative antibi- of vancomycin in patients undergoing Relat Res. 2015;473:2197–2203. otic dose and continuing antibiotics for revision TKA resulted in tissue con- 7. Craig WA. Pharmacokinetic/ fi pharmacodynamic parameters: ratio- 24 hours [12, 13]. This supports centrations of vancomycin ve to 20 nale for antibacterial dosing of mice Burke’s original theory of prophylaxis, times higher than systemic IV admin- and men. Clin Infect Dis.1998;26: which states adequate antibiotic tissue istration despite the lower dose. The 1–10; quiz 11–12. concentrations must be achieved from high tissue concentrations of vanco- 8. Crawford T, Rodvold KA, Solomkin JS. the time of incision to the time of clo- mycin after IORA were maintained Vancomycin for surgical prophylaxis? fi Clin Infect Dis. 2012;54:1474–1479. sure, when contamination is occurring throughout the procedure and the rst 9. de Lalla F, Novelli A, Pellizzer G, [2]. This outcome was clearly attained postoperative day despite a period of Milocchi F, Viola R, Rigon A, Stecca C, in the IORA group in this study despite tourniquet deflation during surgery. Dal Pizzol V, Fallani S, Periti P. Regional

Copyright Ó 2017 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited. 74 Young et al. Clinical Orthopaedics and Related Research® 2017 Knee Society Proceedings

and systemic prophylaxis with teicopla- 18. Mortazavi SMJ, Schwartzenberger J, the infected knee arthroplasty: report nin in monolateral and bilateral total knee Austin MS, Purtill JJ, Parvizi J. Revision from the Swedish Knee Arthroplasty replacement procedures: study of phar- total knee arthroplasty infection: in- Register on 426 surgically revised macokinetics and tissue penetration. cidence and predictors. Clin Orthop Relat cases. Scand J Infect Dis. 2009;41: Antimicrob Agents Chemother. 1993;37: Res. 2010;468:2052–2059. 831–840. 2693–2698. 19. Nickinson RSJ, Board TN, Gambhir AK, 28. Tenover FC, Moellering RCJ. The ratio- 10. de Lalla F, Viola R, Pellizzer G, Lazzarini Porter ML, Kay PR. The microbiology of nale for revising the Clinical and Labo- L, Tramarin A, Fabris P. Regional pro- the infected knee arthroplasty. Int ratory Standards Institute vancomycin phylaxis with teicoplanin in monolateral Orthop. 2010;34:505–510. minimal inhibitory concentration in- or bilateral total knee replacement: an 20. Polk RE, Healy DP, Schwartz LB, Rock terpretive criteria for Staphylococcus open study. Antimicrob Agents Chemo- DT, Garson ML, Roller K. Vancomycin aureus. Clin Infect Dis. 2007;44: ther. 2000;44:316–319. and the red-man syndrome: pharmaco- 1208–1215. 11. Doyon F, Evrard J, Mazas F, Hill C. dynamics of histamine release. J Infect 29. Tobias JD, Ross AK. Intraosseous infu- Long-term results of prophylactic cefa- Dis. 1988;157:502–507. sions: a review for the anesthesiologist zolin versus placebo in total hip re- 21. Rathbone CR, Cross JD, Brown KV, with a focus on pediatric use. Anesth placement. Lancet. 1987;1:860. Murray CK, Wenke JC. Effect of various Analg. 2010;110:391–401. 12. Garcia S, Lozano ML, Gatell JM, Soriano concentrations of antibiotics on osteo- 30. Yamada K, Matsumoto K, Tokimura F, E, Ramon R, SanMiguel JG. Prophylaxis genic cell viability and activity. J Orthop Okazaki H, Tanaka S. Are bone and against infection. Single-dose cefonicid Res. 2011;29:1070–1074. serum cefazolin concentrations ade- compared with multiple-dose cefa- 22. Ritter MA, Barzilauskas CD, Faris PM, quate for antimicrobial prophylaxis? mandole. J Bone Joint Surg Am. 1991;73: Keating EM. Vancomycin prophylaxis Clin Orthop Relat Res. 2011;469: 1044–1048. and elective total joint arthroplasty. Or- 3486–3494. 13. Heydemann JS, Nelson CL. Short-term thopedics. 1989;12:1333–1336. 31. Young SW, Roberts T, Johnson S, Dalton preventive antibiotics. Clin Orthop Relat 23. Rubio-Mart´ınez LM, Lopez-Sanrom´ an´ J, JP, Coleman B, Wiles S. Regional intra- Res. 1986;205:184–187. Cruz AM, Santos M, Andres´ MS, Roman´ osseous administration of prophylactic 14. Hill C, Flamant R, Mazas F, Evrard J. FS. Evaluation of safety and pharmaco- antibiotics is more effective than sys- Prophylactic cefazolin versus placebo in kinetics of vancomycin after intravenous temic administration in a mouse model of total hip replacement. Report of a multi- regional limb perfusion in horses. Am J TKA. Clin Orthop Relat Res. 2015;473: centre double-blind randomised trial. Vet Res. 2005;66:2107–2113. 3573–3584. Lancet. 1981;1:795–796. 24. Scheuch BC, Van Hoogmoed LM, Wilson 32. Young SW, Zhang M, Freeman JT, 15. Lairet J, Bebarta V, Lairet K, Kacprowicz WD, Snyder JR, MacDonald MH, Watson Mutu-Grigg J, Pavlou P, Moore GA. The R, Lawler C, Pitotti R, Bush A, King J. A ZE, Steffey EP. Comparison of intra- Mark Coventry Award: Higher tissue comparison of proximal tibia, distal fe- osseous or intravenous infusion for delivery concentrations of vancomycin with low- mur, and proximal humerus infusion rates of amikacin sulfate to the tibiotarsal joint of dose intraosseous regional versus sys- using the EZ-IO intraosseous device on horses. Am J Vet Res. 2002;63:374–380. temic prophylaxis in TKA: a randomized the adult swine (Sus scrofa) model. Pre- 25. Sierra RJ, Cooney WP, Pagnano MW, trial. Clin Orthop Relat Res. 2014;472: hosp Emerg Care. 2013;17:280–284. Trousdale RT, Rand JA. Reoperations 57–65. 16. McCarthy G, O’Donnell C, O’Brien M. after 3200 revision TKAs: rates, etiology, 33. Young SW, Zhang M, Freeman JT, Vince Successful intraosseous infusion in the and lessons learned. Clin Orthop Relat KG, Coleman B. Higher cefazolin con- critically ill patient does not require Res. 2004;425:200–206. centrations with intraosseous regional a medullary cavity. Resuscitation. 2003; 26. Smith EB, Wynne R, Joshi A, Liu H, prophylaxis in TKA. Clin Orthop Relat 56:183–186. Good RP. Is it time to include vancomy- Res. 2013;471:244–249. 17. Mortazavi SMJ, Molligan J, Austin MS, cin for routine perioperative antibiotic 34. Zhang M, Moore GA, Young SW. De- Purtill JJ, Hozack WJ, Parvizi J. Failure prophylaxis in total joint arthroplasty termination of vancomycin in human following revision total knee arthro- patients? J Arthroplasty. 2012;27:55–60. plasma, bone and fat by liquid plasty: infection is the major cause. Int 27. Stefansd´ ottir´ A, Johansson D, Knutson K, chromatography/tandem mass spectrom- Orthop. 2011;35:1157–1164. Lidgren L, Robertsson O. Microbiology of etry. J Anal Bioanal Tech. 2014;5:196.