Treatment of Periprosthetic Joint Infection in Total Knee Arthroplasty with a Temporary Intramedullary Nail: Is a Long or Short Nail Better?

Keywords

periprosthetic joint infection - total knee arthroplasty - intramedullary nail - two-stage exchange arthroplasty - revision total knee arthroplasty

Periprosthetic joint infection (PJI) after total knee arthroplasty (TKA) is an uncommon but severe complication associated with poor patient outcomes and high expenditures. Although less than 3% of primary TKA patients may develop an infection, up to 23% of revision TKAs are complicated by a subsequent infection.[1] [2] [3] [4] The preferred method of treatment for PJI, two-stage exchange revision arthroplasty, has a success rate of up to 95%.[5] [6] [7] However, its success decreases with each subsequent revision patients undergo, as multiple revisions can lead to significant bone loss and compromised host status.[5] [8] [9] [10] Undergoing revision with traditional antibiotic spacers has proven reliable with adequate eradication rates and functional outcomes, but spacer utilization can lead to bone loss and increased risk of spacer displacement.[11] [12] [13] [14] [15] [16] [17] Therefore, other stabilization and infection eradication techniques have been explored in this cohort of patients.

As a form of static immobilization, intramedullary (IM) nailing have been utilized as an alternative to conventional spacers during two-stage revision. This technique can deliver antibiotics locally and offer sufficient knee stability for patients with preexisting bone loss to allow immediate postoperative ambulation.[18] [19] In the setting of infected TKA, IM nails have demonstrated appropriate efficacy, particularly in patients with extensive bone loss.[20] [21] [22] Specifically, Waldman et al demonstrated that 100% of TKA patients with PJI became infection free after treatment with an IM nail.[20] Similarly, Mohamed et al reported a 74.2% successful treatment rate in repeatedly revised patients with bone loss implanted with a temporary short IM nail.[22] Others have reported positive outcomes with infection-free success rates ranging from 86.5 to 92%.[21] [23] [24] Despite the evidence supporting temporary IM nailing for PJI following TKA, the type of nail that can provide optimal outcomes has not been well defined.

Our institution has utilized both short and long IM nails to treat repeat PJI patients with Anderson Orthopaedic Research Institute (AORI) Type II or III bone loss. To date, no studies have compared the postoperative outcomes of these nails when used in this population to treat PJI during a two-stage revision. Therefore, the aim of this study was to compare short-term postoperative outcomes for patients who underwent temporary long or short IM nail insertion for treatment of PJI. Specifically, we compared: (1) success rates; (2) patient reported and functional outcomes; and (3) complications between patients implanted with a short or a long IM nail following PJI of the knee.

Materials and Methods

Patient Selection

A retrospective chart review was performed for all patients with repeat periprosthetic knee infections at our institution. All patients with Type II or III bone defects, as classified according to the AORI system,[25] who underwent two-stage exchange arthroplasty with a temporary long or short IM nail between March 1, 2010, and June 30, 2018, were included for analysis. Patients with AORI Type 0 or I bone defects, PJI treated with antibiotic suppression, debridement with prosthesis retention, and resection arthroplasty were excluded. Patients with prior operations were not excluded. Demographic factors such as age, sex, body mass index (BMI), race, and health status, were collected for all patients. Our institutional review board determined this project exempt from review as it did not meet the criteria for human subjects research.

A total of 389 patients with an infected TKA were identified between 2009 and 2015. Of this total, 167 patients were treated with antibiotic suppression, irrigation and debridement (I&D), or resection arthroplasty. Of the resulting 222 patients, 155 had no preexisting bone loss and underwent two-stage exchange with articulating or nonarticulating spacers. Thus, a total of 67 patients, 31 men and 36 women, were included, 36 long IM nail patients and 31 short IM nail patients. The patients included in the short nail cohort are part of a previously published study.[22] There were no significant differences in mean ages (63 vs. 62 years; p = 0.746), sex (55.6 vs. 44.4% females; p = 0.193), race (55.6 vs. 61.3% whites; p = 0.635), or mean follow-up (25.4 vs. 26.0 months; p = 0.486) between the long and short IM nail cohorts. Mean BMI was the only patient demographic that significantly differed between the long and short IM nail groups (39.8 vs. 31.8 kg/m²; p < 0.001) ([Table 1]). Regarding health status, we found no difference in mean serum albumin levels (2.9 vs. 3.1; p = 0.283) or McPherson et al's systemic host grade (p = 0.821) between the long and short IM nail cohorts ([Table 2]). Additionally, there was no difference in the organism cultured at initial IM nail placement (p = 0.105) ([Table 3]).

Table 1 Descriptive analysis demonstrating patient demographics

	Long intramedullary nail	Short intramedullary nail	p-Value
Number of patients	36	31
Age[a] (y)	63.4 (9.8)	62.4 (13.8)	0.746
BMI[a] (kg/m2)	39.8 (8.1)	31.8 (7.0)	<0.001
Length of stay[a] (d)	7.4 (7.8)	5.8 (3.2)	0.278
Female	20 (55.6%)	16 (44.4%)	0.193
Race
White	20 (55.6%)	19 (61.3%)	0.635
African American	16 (44.4%)	12 (38.7%)

Abbreviation: BMI, body mass index.

^a Values are reported as mean and standard deviation in parentheses.

Table 2 Host factors

	Long intramedullary nail	Short intramedullary nail	p-Value
Albumin[a]	2.9 (0.6)	3.1 (0.7)	0.283
McPherson et al's systemic host grade
A	3 (8.3%)	4 (12.9%)	0.821
B	25 (69.4%)	20 (64.5%)
C	8 (22.2%)	7 (22.6%)

^a Reported as mean and standard deviation in parentheses.

Table 3 Organism cultured at initial intramedullary nail placement

	Long intramedullary nail	Short intramedullary nail	p-Value
Periprosthetic knee infections (N)	36	31	0.105
No growth	15 (41.7%)	13 (41.9%)
Methicillin-resistant Staphylococcus aureus	4 (11.1%)	4 (12.9%)
Enterococcus group D	2 (5.6%)	1 (3.2%)
Vancomycin-resistant enterococci group D	1 (2.8%)	1 (3.2%)
Pseudomonas aeruginosa	2 (5.6%)	0 (0%)
Staphylococcus aureus	4 (11.1%)	0 (0%)
Coagulase-negative staphylococci	4 (11.1%)	8 (25.8%)
Streptococcus anginosus	1 (2.8%)	0 (0%)
Streptococcus viridians	1 (2.8%)	0 (0%)
Candida albicans	0 (0%)	1 (3.2%)
Escherichia coli	0 (0%)	1 (3.2%)
Polymicrobial	1 (2.8%)	2 (6.5%)
Proteus mirabilis	1 (2.8%)	0 (0%)

Management Technique

Surgery consisted of removal of the infected prostheses, I&D, and IM nail insertion. The IM nails were selected based on surgeon preference and were coated with antibiotic cement. The cement was composed of liquid monomer, 3.6 g of tobramycin, and 1 g of vancomycin per every 40-g package of cement prior to insertion. Tobramycin antibiotic powder (4.6 g) alone was used if a patient had an allergy to vancomycin. All patients received standardized institutional postoperative care and were instructed to weight-bear as tolerated immediately. Parenteral antibiotics were prescribed for at least 6 weeks under the guidance of blood and aspirate cultures as well as microbiome sensitivities. Antibiotics were prolonged in cases where infection was not eradicated. Serum infection markers, such as C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR), were assessed throughout the course of antibiotic therapy. In cases where the CRP and ESR levels were questionable, joint aspiration was performed to rule out ongoing infection. Once patients were considered to have cleared the infection by improvement in symptoms, downward trending in CRP and ESR levels, or a negative aspiration, they were scheduled for reimplantation. Patients were reimplanted if intraoperative pathology reported negative frozen sections.

Variables

Health status was graded preoperatively according to patient serum albumin levels and the McPherson et al's classification.[26] The microorganism cultured from intraoperative tissue specimens was collected and compared between the cohorts. Treatment success was determined using the Delphi-based consensus definition of successfully treated PJI: infection eradication, no further surgical intervention for infection after reimplantation, and no PJI-related mortality.[27]

Postoperative complications such as reinfection and amputation were documented. Patient-reported outcomes included pain and satisfaction. Perioperative pain intensity was measured by utilizing patient visual analog scale (VAS) scores. In addition to preoperative baseline scores, VAS scores were collected up to 48 hours postoperatively at 8-hour intervals and reported as area under the curve. Satisfaction was assessed via a postoperative survey, and scored out of 10. Functional outcomes were assessed using VAS and Knee Society scores (KSSs) from the patient's last clinic follow-up visit.

Data Analysis

Continuous and categorical variables were assessed using Student's t-test, Mann–Whitney U test, and chi-squared test, respectively. Logistic regression analyses were conducted to assess the effect of IM nail length on success rate while adjusting for age, sex, BMI, and race. A p-value of 0.05 was set as the threshold for statistical significance. All statistical analysis was performed using SPSS version 25 (IBM Corporation; Armonk, NY).

Results

Success Rate

After adjusting for age, sex, BMI, and race, there was no differences in success rates for reimplanted patients treated with long and short IM nails (odds ratio 0.992; p = 0.847) at a mean follow-up of 26.0 months ([Table 4]). Fewer patients with a long IM nail went on to reimplantation compared with patients with a short IM nail (52.8 vs. 83.9%; p = 0.007) ([Table 5]). In the interchange period, there were no significant differences between long and short IM nail cohorts requiring exchange for a second nail (31.6 vs. 11.5%; p = 0.137).

Table 4 Logistic regression model for treatment success

Variable	Odds ratio	p-Value	95% confidence interval
Age	0.997	0.917	0.945–1.053
BMI	0.992	0.847	0.915–1.076
Race
Black or African American	–	Reference	–
White	3.354	0.033	1.101–10.223
Sex
Female	–	Reference	–
Male	1.472	0.499	0.480–4.510
Host immune status
A	–	Reference	–
B	1.234	0.827	0.189–8.071
C	0.907	0.934	0.90–9.156
Albumin	1.206	0.724	0.427–3.402
Required nail exchange	0.512	0.422	0.100–2.619
IM nail length
Long	–	Reference	–
Short	0.992	0.847	0.915–1.076

Abbreviations: BMI, body mass index; IM, intramedullary.

Table 5 Outcomes and complications for temporary long and short intramedullary nails

	Long intramedullary nail	Short intramedullary nail	p-Value
Number of periprosthetic knee infections (N)	36	31
Outcome measure
Follow-up for all[a] (mo)	31.5 (24.0–226.0)	32.0 (24.2–85.0)	0.806
Duration of antibiotics[a] (d) (SD)	42.0 (35–42)	42.0 (14–56)	0.457
LOS[b] (d)	7.42 (7.8)	5.77 (3.2)	0.278
Reimplantation
Reimplanted	19 (52.8%)	26 (83.9%)	0.007
Time until reimplant[a] (mo)	10.3 (4.0–41.0)	2.7 (0.1–16)	0.008
Treatment success	16 (84.2%)	19 (73.1%)	0.481
Interchange period
Required a second intramedullary nail[c]	6 (31.6%)	3 (11.5%)	0.137
Complications following reimplantation
Infection[c]	3 (15.8%)	3 (11.5%)	0.679
Amputation[c]	0 (0%)	2 (7.7%)	0.210
Retention of nail
Retention of nail[c]	17 (47.2%)	5 (16.1%)	0.007
Treatment success[c]	16 (94.1%)	4 (80.0%)	0.334
Complications with retention of nail
Infection[c]	1 (5.9%)	0 (0%)	0.999
Amputation[c]	1 (5.9%)	0 (0%)	0.999
Irrigation and debridement[c]	0 (0%)	1 (20.0%)	0.999

Abbreviations: LOS, length of stay; SD, standard deviation.

^a Median (range).

^b The values are given as the mean and range in parentheses.

^c The values are given as the number of cases with the percentage in parentheses. The percentages were determined from the total number of cases that retained the intramedullary nail and cases that went on to reimplantation.

Patient-Reported and Functional Outcomes

There were no differences in pain intensity between the long and short IM nail groups during the 48-hour interval postoperatively (114.61 vs. 134.93 through 0–24 hours; p = 0.148 and 128.50 vs. 143.74 points through 24–48 hours; p = 0.191) ([Table 6]). Similarly, there were no differences reported in postoperative satisfaction (7.86 vs. 7.68; p = 0.515). There were also no differences in VAS pain scores (3.39 vs. 4.45 points; p = 0.126) and KSS functional outcome scores (150.61 vs. 166.26 points; p = 0.117) between patients reimplanted with a long or short IM nail at a mean follow-up of 32 months.

Table 6 Patient-reported outcomes and functional scores for patients treated with temporary long or short intramedullary nails

	Long intramedullary nail	Short intramedullary nail	p-Value
Pain intensity
Pain intensity (AUC) 0–24 h	114.61 (44)	134.93 (65.13)	0.148
Pain intensity (AUC) 24–48 h	128.50 (38.21)	143.74 (53.37)	0.191
Patient-reported satisfaction (out of 10)
Satisfaction	7.86 (1.15)	7.68 (1.13)	0.515
Functional and PRO scores at last follow-up for patients who retained the IM nail
KSS	117.61 (27.91)	103.52 (23.41)	0.198
VAS	4.52 (2.63)	5.29 (3.16)	0.154
Satisfaction	7.65 (1.11)	7.33 (1.03)	0.553
Functional and PRO scores at last follow-up for reimplanted patients
KSS	150.61 (45.01)	166.26 (33.72)	0.117
VAS	3.39 (2.23)	4.45 (3.19)	0.126
Satisfaction	8.05 (1.18)	7.76 (1.16)	0.416

Abbreviations: AUC, area under the curve; IM, intramedullary; KSS, Knee Society score; PRO, patient-reported outcomes; VAS, visual analog scale.

Note: The values are given as the mean and standard deviation in parentheses.

Complications

Following reimplantation, there was no difference in the number of patients who became reinfected (15.8 vs. 11.5%; p = 0.679) or went on to amputation (0 vs. 7.7%; p = 0.210) ([Table 5]).

Discussion

PJI following TKA is a serious complication associated with increased patient morbidity and mortality. Temporary IM nail utilization in combination with two-stage exchange arthroplasty is an option in the treatment of PJI when severe bone loss is present, as the nail offers sufficient knee stability for immediate mobilization.[18] [22] Following infection eradication and explantation of the nail, patient function can potentially be restored, and activity levels may improve. Both long and short IM nails have been successful in two-stage exchange, but their outcomes have not been previously compared. Therefore, this study compared short-term postoperative outcomes for patients who underwent long or short IM nail insertion for treatment of PJI during two-stage exchange arthroplasty. Our findings demonstrated fewer long nail patients underwent reimplantation, retaining their nail after surgery. We also found that there were no statistical differences in reinfection, amputation, or successful infection eradication rates between the cohorts. Additionally, no differences in patient-reported or functional outcomes were noted between the cohorts. The utilization of both long and short IM nails during a two-stage revision TKA are comparable in patients with prior bone loss.

This study is not without limitations. We found a difference in BMI between the long and short IM nail cohorts. Long IM nails are implanted in patients with larger stature due to surgeon preference, and therefore, it would be expected to see a higher BMI in this group. Nevertheless, this difference did not affect the statistical analysis as this variable was adjusted for in the logistic regression analysis. Furthermore, as the incidence of PJI is as low as 0.7 to 2%,[28] a large number of study patients are necessary to detect significant differences between cohorts. Although this study is likely underpowered to detect this difference, a comparison between long and short IM nails has not been explored, and is therefore valuable, particularly because differing implants can achieve temporary fixation during PJI treatment. Furthermore, post hoc power analysis with study parameters of 80% power and an α = 0.05 revealed the need for 68 total patients, which is in line with our study.

Our study found that fewer patients with a long IM nail went on to reimplantation. Per chart review, patients felt satisfied that they had adequate stability and pain control and were reluctant to undergo further surgical intervention. These patients were in effect, successfully managed with a one-stage exchange. Of those who underwent reimplantation, we found varied success rates for infection eradication between the long and short nail cohorts. Similarly, studies regarding nonarticulating cement spacers, which are comparable to IM nails, report wide-ranging infection eradication rates. Fehring et al performed a retrospective study on 25 PJI patients treated with a static cement spacer and found 22 patients (88%) were free of infection at a mean follow-up of 36 months.[16] The authors reported unexpected bone loss in 15 (60%) of the 25 patients, a complication commonly associated with the use of these spacers.[4] [13] In another retrospective review of failed infected TKAs, Johnson et al reported a success rate of 83% (28 of 34 patients) in their cohort treated with nonarticulating cement spacers.[12] Conversely, Choi et al performed a retrospective study on 33 PJI patients treated with static cement spacers and found 22 patients (67%) were infection free at 58 months of follow-up.[29] These studies demonstrate the variability of infection eradication with traditional static spacers, which help corroborate the eradication rates of the IM nails utilized as nontraditional spacers in this study.

We found no differences in postoperative pain intensity or KSSs between the groups at a mean 2-year follow-up. Several studies have shown that functional outcome also improved when patients treated with static spacers undergo reimplantation. Lichstein et al performed a retrospective case series on 107 infected TKA patients and found that KSS knee scores improved from a preoperative median of 36 (range: 24–48) to a postoperative median of 86 (range: 22–45).[30] They also found that KSS function scores increased from a median of 32 (range: 22–45) to a postoperative median of 85 (range: 61–97). Fehring et al utilized the Hospital for Special Surgery rating to evaluate functional outcomes and found that patients treated with a static spacer block (n = 25) had an improved mean score of 83 points (range: 37–98).[16] Our study expands on these previous studies by including a pain intensity analysis, which these studies fail to include.

Our findings demonstrated no difference in the number of patients who became reinfected or went on to amputation. In the cohort of previously published short nail patients, Mohamed et al reported the observed overall success rate of 74.1%, with a reinfection rate of 11.5% and no amputations.[22] Other studies report similar reinfection rates and subsequent surgical management. A retrospective review by Emerson et al studied postoperative outcomes for 26 patients treated with a static block spacer for infected TKAs.[17] The authors reported a reinfection rate of 7.6% at 36 months that increased to 23% (6 of 26) after an average of 7.5 years. In a retrospective study of 38 patients implanted with static spacers, Freeman et al reported three reoperations (7.8%) in the group of patients who remained infected or became reinfected.[31] In another retrospective study evaluating outcomes for cement spacers (n = 10), Jämsen et al reported that one patient underwent an above-the-knee amputation due to a life-threatening infection.[32] Despite the small sample sizes reported, results from these studies coincide with our findings, reporting low reinfection and amputation rates. Our comparative study demonstrates similar outcomes between IM nails and static spacers, suggesting they have a place in the treatment of infected TKA during two-stage exchange arthroplasty. A long IM nail offers the potential for a one-stage revision, while a short IM nail offers increased mobilization during the interim of a two-stage exchange. This information can provide surgeons with more tools to assist their more complex patients.

Conclusion

Temporary IM nail placement can be an effective alternative technique to traditional antibiotic spacers for PJI patients with considerable bone loss. This study compared long and short IM nails in this population of patients, and found a subset of patients implanted with a long IM nail were successfully treated with one-stage exchange. We also found no difference in success rates for infection eradication for reimplanted patients treated with a long or short IM nail at short-term follow-up. These data suggest that the use of long and short IM nails during two-stage exchange can have equal utility in PJI patients with severe bone defects.

• References

• 1 Peersman G, Laskin R, Davis J, Peterson M. Infection in total knee replacement: a retrospective review of 6489 total knee replacements. Clin Orthop Relat Res 2001; (392) 15-23
• 2 Phillips JE, Crane TP, Noy M, Elliott TSJ, Grimer RJ. The incidence of deep prosthetic infections in a specialist orthopaedic hospital: a 15-year prospective survey. J Bone Joint Surg Br 2006; 88 (07) 943-948
• 3 Pulido L, Ghanem E, Joshi A, Purtill JJ, Parvizi J. Periprosthetic joint infection: the incidence, timing, and predisposing factors. Clin Orthop Relat Res 2008; 466 (07) 1710-1715
• 4 Pivec R, Naziri Q, Issa K, Banerjee S, Mont MA. Systematic review comparing static and articulating spacers used for revision of infected total knee arthroplasty. J Arthroplasty 2014; 29 (03) 553-7.e1
• 5 Stammers J, Kahane S, Ranawat V. et al. Outcomes of infected revision knee arthroplasty managed by two-stage revision in a tertiary referral centre. Knee 2015; 22 (01) 56-62
• 6 Mortazavi SMJ, Schwartzenberger J, Austin MS, Purtill JJ, Parvizi J. Revision total knee arthroplasty infection: incidence and predictors. Clin Orthop Relat Res 2010; 468 (08) 2052-2059
• 7 Silvestre A, Almeida F, Renovell P, Morante E, López R. Revision of infected total knee arthroplasty: two-stage reimplantation using an antibiotic-impregnated static spacer. Clin Orthop Surg 2013; 5 (03) 180-187
• 8 Tigani D, Trisolino G, Fosco M, Ben Ayad R, Costigliola P. Two-stage reimplantation for periprosthetic knee infection: influence of host health status and infecting microorganism. Knee 2013; 20 (01) 9-18
• 9 Calton TF, Fehring TK, Griffin WL. Bone loss associated with the use of spacer blocks in infected total knee arthroplasty. Clin. Orthop. Relat. Res. 1997; LLC: 148-154
• 10 Fehring KA, Abdel MP, Ollivier M, Mabry TM, Hanssen AD. Repeat two-stage exchange arthroplasty for periprosthetic knee infection is dependent on host grade. J Bone Joint Surg Am 2017; 99 (01) 19-24
• 11 Romanò CL, Gala L, Logoluso N, Romanò D, Drago L. Two-stage revision of septic knee prosthesis with articulating knee spacers yields better infection eradication rate than one-stage or two-stage revision with static spacers. Knee Surg Sports Traumatol Arthrosc 2012; 20 (12) 2445-2453
• 12 Johnson AJ, Sayeed SA, Naziri Q, Khanuja HS, Mont MA. Minimizing dynamic knee spacer complications in infected revision arthroplasty. Clin Orthop Relat Res 2012; 470 (01) 220-227
• 13 Mazzucchelli L, Rosso F, Marmotti A, Bonasia DE, Bruzzone M, Rossi R. The use of spacers (static and mobile) in infection knee arthroplasty. Curr Rev Musculoskelet Med 2015; 8 (04) 373-382
• 14 Park S-J, Song E-K, Seon J-K, Yoon T-R, Park G-H. Comparison of static and mobile antibiotic-impregnated cement spacers for the treatment of infected total knee arthroplasty. Int Orthop 2010; 34 (08) 1181-1186
• 15 Hsu YC, Cheng HC, Ng TP, Chiu KY. Antibiotic-loaded cement articulating spacer for 2-stage reimplantation in infected total knee arthroplasty: a simple and economic method. J Arthroplasty 2007; 22 (07) 1060-1066
• 16 Fehring TK, Odum S, Calton TF, Mason JB. Articulating versus static spacers in revision total knee arthroplasty for sepsis. The Ranawat Award. Clin Orthop Relat Res 2000; (380) 9-16
• 17 Emerson Jr RH, Muncie M, Tarbox TR, Higgins LL. Comparison of a static with a mobile spacer in total knee infection. Clin Orthop Relat Res 2002; (404) 132-138
• 18 Wood JH, Conway JD. Advanced concepts in knee arthrodesis. World J Orthop 2015; 6 (02) 202-210
• 19 Wiedel JD. Salvage of infected total knee fusion: the last option. Clin Orthop Relat Res 2002; (404) 139-142
• 20 Waldman BJ, Mont MA, Payman KR. et al. Infected total knee arthroplasty treated with arthrodesis using a modular nail. Clin Orthop Relat Res 1999; (367) 230-237
• 21 Bargiotas K, Wohlrab D, Sewecke JJ, Lavinge G, DeMeo PJ, Sotereanos NG. Arthrodesis of the knee with a long intramedullary nail following the failure of a total knee arthroplasty as the result of infection. Surgical technique. J Bone Joint Surg Am 2007; 89 (Suppl 2 Pt.1): 103-110
• 22 Mohamed NS, Etcheson JI, Wilkie WA. et al. Two-stage exchange using a short intramedullary nail for treatment of periprosthetic knee infections: a technique worth questioning. J Knee Surg 2020; (ePub ahead of print)
• 23 Friedrich MJ, Schmolders J, Wimmer MD. et al. Two-stage knee arthrodesis with a modular intramedullary nail due to septic failure of revision total knee arthroplasty with extensor mechanism deficiency. Knee 2017; 24 (05) 1240-1246
• 24 Gathen M, Wimmer MD, Ploeger MM. et al. Comparison of two-stage revision arthroplasty and intramedullary arthrodesis in patients with failed infected knee arthroplasty. Arch Orthop Trauma Surg 2018; 138 (10) 1443-1452
• 25 Qiu YY, Yan CH, Chiu KY, Ng FY. Review article: bone defect classifications in revision total knee arthroplasty. J Orthop Surg (Hong Kong) 2011; 19 (02) 238-243
• 26 McPherson EJ, Woodson C, Holtom P, Roidis N, Shufelt C, Patzakis M. Periprosthetic total hip infection: outcomes using a staging system. Clin Orthop Relat Res 2002; (403) 8-15
• 27 Diaz-Ledezma C, Higuera CA, Parvizi J. Success after treatment of periprosthetic joint infection: a Delphi-based international multidisciplinary consensus. Clin Orthop Relat Res 2013; 471 (07) 2374-2382
• 28 Kubista B, Hartzler RU, Wood CM, Osmon DR, Hanssen AD, Lewallen DG. Reinfection after two-stage revision for periprosthetic infection of total knee arthroplasty. Int Orthop 2012; 36 (01) 65-71
• 29 Choi H-R, Malchau H, Bedair H. Are prosthetic spacers safe to use in 2-stage treatment for infected total knee arthroplasty?. J Arthroplasty 2012; 27 (08) 1474-1479.e1
• 30 Lichstein P, Su S, Hedlund H. et al. Treatment of periprosthetic knee infection with a two-stage protocol using static spacers. Clin Orthop Relat Res 2016; 474 (01) 120-125
• 31 Freeman MG, Fehring TK, Odum SM, Fehring K, Griffin WL, Mason JB. Functional advantage of articulating versus static spacers in 2-stage revision for total knee arthroplasty infection. J Arthroplasty 2007; 22 (08) 1116-1121
• 32 Jämsen E, Sheng P, Halonen P. et al. Spacer prostheses in two-stage revision of infected knee arthroplasty. Int Orthop 2006; 30 (04) 257-261