Abstract
Purpose
Late acute haematogenous total knee arthroplasty (LAH TKA) infections represent approximately 20% of all prosthetic infections. Debridement, antibiotics, and implant retention (DAIR) is a widely accepted treatment; however, it carries a significant risk of failure. The study aims to identify risk factors for DAIR failure in LAH TKA infections, focusing on blood cultures (BC) as a potential contributor.
Methods
A retrospective cohort study was conducted on 37 cases of LAH TKA infections from 2015 to 2023. Patients were divided into two groups: 20 in the success group (SG) and 17 in the failure group (FG). The study analyzes various factors, including demographics, comorbidities, serum and joint fluid biomarkers, blood and intraoperative cultures, prior infection or antibiotic use, and surgical and post-surgical variables.
Results
Positive BC were more frequent in the FG compared with the SG (P = 0.03), and were also more common in patients with a history of prior infection (P = 0.03). Logistic regression identified positive BC as the only significant predictor of DAIR failure (odds ratio (OR) 12, 95% confidence interval (CI) 1.1–18, P = 0.04), even after adjusting for other variables. Positive intraoperative cultures were more frequent in the FG, particularly in those with a prior infection history (P = 0.08), even if they were receiving antibiotics (P = 0.05).
Conclusions
Positive BC may be an additional risk factor for DAIR failure in LAH TKA. This highlights the importance of routinely performing BC during diagnosis.
Level of evidence: level III.
Keywords: Arthroplasty, Infection, Blood culture, Culture, DAIR, Clinical microbiology
Introduction
Prosthetic joint infections (PJIs) are a serious complication for orthopedic surgeons, associated with high morbidity and even mortality [1–3]. Although the incidence of PJI is currently low, it is expected to rise in the coming years [4].
PJIs have been typically classified according to Zimmerli’s classification into four types: immediate post-surgery, chronic, late acute hematogenous (LAH), and positive intraoperative cultures [5]. LAH PJI is defined as an infection that occurs from hematogenous spread to a previously well-functioning and pain-free prosthesis. This type of infection often occurs months or even years after a total knee arthroplasty (TKA) [5, 6]. LAH infections account for approximately 20–35% of all PJIs, though this percentage varies depending on the series [7, 8]. In many cases, the source of infection remains unidentified, and owing to a lack of consensus on its definition, the true incidence of hematogenous infections is still unknown [7, 8].
The most common sources of microorganisms that spread through the bloodstream include the skin and soft tissues, the cardiovascular system, the oral cavity, and the gastrointestinal or urogenital tracts [7, 9]. The most frequently isolated microorganisms are Staphylococcus aureus and Streptococcus spp. [7, 9].
Treatment of LAH PJIs depends on factors such as the duration of symptoms, the condition of the implants (weather well-fixed or loose), and the type of microorganism involved [1, 10]. One of the most widely accepted treatment options is debridement, antibiotics, and implant retention (DAIR), as it aims to maintain a functional and stable implant, making it a preferred choice among surgeons [1, 10, 11].
The success rate of DAIR in acute infections is generally high (70–90%), whereas, in hematogenous infections, this rate drops significantly to around 20–60% [12, 13]. The reason for this difference remains unclear [8]. Several risk factors for DAIR failure in LAH PJI have been identified, and a scoring system known as CRIME80 has been proposed. Key risk factors include: the causative microorganism, with Staphylococcus aureus being particularly problematic; failure to perform a polyethylene exchange during surgery and comorbid factors such as rheumatoid arthritis, chronic obstructive pulmonary disease (COPD), C-reactive protein (CRP) levels above 150 mg/L, male sex , age over 80 years, and intraarticular fracture as the indication for the prosthesis [11].
However, the impact of positive blood cultures on DAIR outcomes and prognosis is less well understood. Few studies have explored this relationship [14, 15]. Klement et al. included all types of PJI and various surgical treatments in their study and found that patients with positive blood cultures tended to have worse outcomes [14]. Similarly, Kuo et al. focused specifically on hematogenous infections, including both hip and knee prosthetic infections, and reported a correlation between positive blood cultures and poorer outcomes [15].
The aim of our study is to identify factors that contribute to DAIR failure in late acute hematogenous total knee arthroplasty (LAH TKA) infections, with a specific focus on positive blood cultures. We hypothesize that a higher bacterial load may trigger a more intense inflammatory response, increase the likelihood of positive blood and intraoperative cultures, and decrease the chances of treatment success. We further propose that positive blood culture will be an independent risk factor for poor prognosis in DAIR treatment of LAH TKA infections, distinct from other known risk factors.
Materials and methods
A retrospective cohort study was conducted involving 50 patients with LAH TKA infections treated with DAIR surgery between January 2015 and December 2023. The diagnosis of late acute hematogenous (LAH) infection was based on the Zimmerli classification, which defines hematogenous infections as those occurring acutely in a previously well-functioning prosthesis more than 3 months after implantation. In addition, the diagnosis was supported by clinical signs of acute infection (such as sudden onset of joint pain, erythema, or swelling), elevated inflammatory markers (CRP and ESR), and microbiological confirmation through positive synovial fluid and/or intraoperative cultures [5]. DAIR failure was defined as the inability to fully eradicate the infection, necessitating additional intervention such as repeat DAIR or revision surgery, the need for suppressive antibiotic therapy, or infection-related mortality, in accordance with the Delphi International Consensus criteria [16]. Failure could be diagnosed at any point during the follow-up period.
Of the 50 patients, 13 were excluded owing to the absence of blood cultures prior to the surgery. The remaining patients were divided into two groups on the basis of surgical outcome: 20 in the successful group (SG) and 17 in the failure group (FG). Among the 17 patients in the failure group, the primary cause was the inability to eradicate the infection (82.4%). This led to the need for either a new intervention (70.6%) or continued suppressive antibiotic therapy (11.8%). Infection-related mortality accounted for 17.6% of the cases.
The study was approved by the Institutional Review Board (PI-23-117). The authors undertake to comply with the fundamental ethical principles contained in the Declaration of Helsinki (Helsinki, October 2024). The data for this study were obtained from an anonymized database provided by the Information Technology Department, as informed consent was not available since this was a retrospective study of previous years.
Study variables
Various demographic variables were collected, including age, sex, affected side, and body mass index (BMI). Patient comorbidities considered potential risk factors for DAIR failure were also recorded, including diabetes, chronic kidney disease (CKD), chronic obstructive pulmonary disease (COPD), and rheumatoid arthritis (RA). Additionally, the CRIME80 score was calculated for all patients.
Laboratory data, including serum leukocyte counts, CRP serum levels, and joint fluid cell count, percentage of polymorphonuclear (PMN), as well as glucose and protein levels, were collected. Inquiry was made regarding the presence of any previous reported infection (PRI), as well as the need for recent antibiotic therapy.
Surgical variables analyzed included the primary diagnosis, type of TKA surgery (primary or revision), surgeon expertise (with “expert” defined as a surgeon with over 5 years of experience in knee surgery or performing 80 TKA per year), polyethylene exchange, and intraoperative soft tissue culture results. Post-surgical data, including the duration of antibiotic treatment, were also examined.
Blood cultures
Blood cultures (BC) were taken in the emergency department, prior to surgical treatment, in all patients with suspected bacteremia. All blood cultures were drawn according to hospital protocol, following three steps. First disinfection, the venepuncture site was disinfected using an alcohol antiseptic solution, ensuring the area was not touched by hand afterward. Second, the blood was drone using syringe. Third, aerobic and anaerobic blood cultures bottles (BD BACTEC™, Becton, Dickinson and Company, Franklin Lakes, NJ) were filled with 8–10 ml of blood each, ensuring the bottles were held upright. The blood samples were promptly sent to the microbiology department for processing.
Surgical treatment
All surgeries were performed by orthopedic surgeons specialized in prosthetic joint infections using a standardized surgical technique as soon as clinically feasible.
Antibiotic prophylaxis with 2 g of cefazolin was administered when culture samples were taken. For patients allergic to penicillin, 900 mg of clindamycin was used instead.
The DAIR procedure was conducted under general anesthesia and ischemia. The tourniquet was inflated after 3 min of elevation without expression of the lower leg. An iterative anterior knee approach was made, following the previous scar. Joint fluid aspiration was performed prior to the medial arthrotomy and was sent for culture, cell count, and biochemical testing. Key surgical steps included polyethylene exchange, complete synovectomy, extensive debridement including the posterior compartment, and irrigation with 9 L of sterile normal saline. After irrigation, the superficial drape layer of the surgical field was changed and a new polyethylene was implanted. A minimum of six intraoperative cultures (IC) samples were taken from the most affected areas: the synovia in the suprapatellar pouch, the medial and lateral gutters, the intercondylar notch, the Hoffa fat pat, the posterior capsule. Additionally, sonication of the polyethylene was performed. The tourniquet was deflated before wound closure, which was done in layers. The deep layer was closed using #2 Vicryl® (90% de glycolide y 10% de l-lactide), #2/0 was used for the subcutaneous layer, and staples were applied for the skin. An intra-articular drain was always used for 24–48 h.
Antibiotic treatment
After culture samples were collected, empiric antibiotic therapy was initiated using a broad-spectrum antibiotic regimen of IV ceftazidime (2 g/8 h) and vancomycin (1 g/12 h), despite knowing the preoperative pathogen. A peripherally insertable central catheter was placed for antibiotic administration. Once the microorganism was identified or confirmed through blood and intraoperative cultures, the antibiotic regime was adjusted to a narrower-spectrum antibiotic on the basis of antibiotic sensitivity testing. By day 7 post-surgery, if the patient showed good clinical progress and negative blood cultures, the IV antibiotics were switched to oral antibiotics, which were continued for up to 8 weeks. In patients with positive blood cultures, the IV antibiotics were extended for an additional 2 weeks. If there was clinical improvement by day 14, the IV antibiotics were then transitioned to oral administration, also continued for up to 8 weeks.
Hospitalization
All patients received antithrombotic prophylaxis with 40 mg (4000 UI) of low-molecular-weight heparin for 30 days post-surgery. The postoperative rehabilitation protocol was standardized for all patients. At 24 h post-surgery, the patients were encouraged to sit and begin ambulation as tolerated. The second postoperative day, passive joint mobilization including machine-assisted mobilization using kinetec® was started. On the third postoperative day, patients participated in physiotherapy session at the hospital gym. If recovery progress well, patients were discharged home after the seventh postoperative day, supported by a home hospitalization service.
Follow-up
Outpatient follow-ups were managed by a multidisciplinary team that included orthopedic surgeons and infectious disease specialists. Each visit involved blood test and radiographs to monitor recovery. The first visit occurred 1 month after discharge at home. A second visit was scheduled after completing the antibiotic treatment. Additional follow-up visits were scheduled at 3, 6, and 12 months post-surgery. If the patient showed good progress after the first year, annual visits were arranged.
Statistical analysis
The Shapiro–Wilk test was used to assess the normality of variable distributions. Continuous parametric data are expressed as mean ± standard deviation (SD), while nonparametric data are reported as median and interquartile range (IQR). Categorical data are presented as percentages.
Student’s t-test was used for the analysis of parametric data. Mann–Whitney U test was applied for nonparametric variables. The chi-squared test was used to analyze categorical data, with Fisher’s exact test applied when expected values in any category were fewer than five. The Haldane–Anscombe correction for the odds ratio was applied in cases where any cell had a value of zero.
To evaluate factors associated with DAIR failure, logistic regression analyses were performed with DAIR outcome (success versus failure) as the dependent variable. On the basis of clinical relevance and data availability, the CRIME80 score, causative microorganism, blood culture positivity, polyethylene exchange, and duration of antibiotic treatment were considered as independent variables. Intraoperative culture results were excluded from the multivariable analysis owing to their strong correlation with DAIR outcome and the risk of collinearity. Likewise, the indication for total knee arthroplasty was not included in the final model because of its limited variability and lack of independent association with the outcome. Given the limited number of failure events, a parsimonious multivariable model was constructed to reduce the risk of overfitting and to allow a more robust estimation of the independent effects of the selected variables.
All analyses were performed using the Statistical Package for the Social Sciences (SPSS® version 24.0; IBM Corp, Armonk, NY, USA) for Windows. Statistical significance was set at p < 0.05.
Results
Patient characteristics (Table 1)
Table 1.
Description of patient’s characteristics and comorbidities according to groups
| Variable | DAIR outcome | P- value | ||
|---|---|---|---|---|
| Successful (n = 20) | Failure (n = 17) | |||
| Age (years) | 67.8 ± 14.1 | 74.2 ± 9.5 | 0.13 | |
| BMI (kg/cm2) | 33.2 ± 5.2 | 31.4 ± 5.3 | 0.32 | |
| COPD (%) | 66.6 | 33.3 | 0.29 | |
| RA (%) | 50.0 | 50.0 | 0.80 | |
| Diabetes (%) | 57.1 | 42.9 | 0.77 | |
| Ischemic heart disease (%) | 52.6 | 47.4 | 0.86 | |
| CKD (%) | 30.0 | 70.0 | 0.08 | |
| HTN (%) | 48.4 | 51.6 | 0.12 | |
| Sex (%) | Female | 63.0 | 37.0 | 0.13 |
| Male | 30.0 | 70.0 | ||
| Side (%) | Right | 50.0 | 50.0 | 0.67 |
| Left | 56.5 | 43.5 | ||
| CRIME80 score | 0 ± 2.3 | 1 ± 2 | 0.09 | |
Results displayed as mean ± standard deviation, percentages, or absolute value
BMI, body mass index; COPD, chronic obstructive pulmonary disease; RA, rheumatoid arthritis; CKD, chronic kidney disease; HTN, hypertension
*P-value < 0.05
The SG and FG groups were comparable in terms of age, BMI, laterality, comorbidities (COPD, RA, diabetes, CKD, HTN), and CRIME80 score. However, men were more prevalent in the FG, while women were more frequent in the SG (P = 0.13).
Serum and Joint fluid laboratory biomarkers (Table 2)
Table 2.
Laboratory biomarkers, blood, and intraoperative culture results according to groups
| Variable | DAIR outcome | P- value | ||
|---|---|---|---|---|
| Successful (n = 20) | Failure (n = 17) | |||
| Blood tests | Leukocytes (× 109L) | 11945 ± 4234 | 12113 ± 5093 | 0.91 |
| CRP levels (mg/L) | 138.0 ± 113 | 207.0 ± 118 | 0.09 | |
| Joint aspiration | Cells (mm3) | 32138.2 ± 21634 | 34805.4 ± 22115 | 0.73 |
| PMN (%) | 90.6 ± 7.5 | 88.5 ± 7.9 | 0.45 | |
| Glucose (mg/dL) | 51.8 ± 36.6 | 66.8 ± 79.3 | 0.48 | |
| Proteins (g/L) | 43.5 ± 8.6 | 40.7 ± 7.2 | 0.33 | |
| Blood cultures, n (%) | Positive | 2 (10.0) | 7 (41.2) | 0.03* |
| Negative | 18 (90.0) | 10 (58.8) | ||
| Blood culture microorganisms, n (%) | S. aureus | 0 (0.0) | 4 (23.5) | 0.44 |
| CoNS | 1 (5.0) | 0 (0.0) | ||
| Streptoccoccus spp. | 0 (0.0) | 1 (5.9) | ||
| GNB | 1 (5.0) | 2 (11.8) | ||
| Intraoperative cultures, n (%) | Positive | 10 (50.0) | 14 (82.4) | 0.08 |
| Negative | 10 (50.0) | 3 (17.6) | ||
| Intraoperative culture microorganisms, n (%) | S. aureus | 4 (20.0) | 8 (47.1) | 0.32 |
| CoNS | 2 (10.0) | 1 (5.9) | ||
| Streptococcus spp | 1 (5.0) | 1 (5.9) | ||
| GNB | 3 (15.0) | 4 (23.5) | ||
Results are displayed as mean ± standard deviation, percentages, or absolute value
CRP, C-reactive protein; PMN, polymorphonuclear cells; S. aureus, Staphylococcus aureus; CoNS, coagulase-negative species; GNB, Gram-negative bacilli
*P-value < 0.05
CRP serum levels were higher in the FG (207.0 ± 118 mg/L) compared with the SG (138.0 ± 113.0 mg/L) (P = 0.09). Serum leukocyte counts and joint aspiration cell count, PMN percentage, glucose, and proteins levels were similar between the groups (n.s).
Absence of blood cultures
To evaluate potential selection bias due to missing blood cultures (13/50, 26%), we performed sensitivity analyses. First, we repeated the primary analysis after reclassifying patients without blood cultures as “negative blood culture” and including them in the analysis. Second, we tested an extreme-case scenario assuming that all patients without blood cultures would have had positive blood cultures. The association between positive blood cultures and DAIR failure remained robust when missing cases were considered negative (relative risk, RR 2.45; 95% CI 1.39–4.33), and it was attenuated only under the extreme assumption that all missing cases were positive (RR 1.27; 95% CI 0.65–2.50).”
Blood and intraoperative cultures (Table 2)
A total of 37 blood cultures were obtained prior to surgery: 20 from the SG and 17 from the FG. Nine of the blood cultures were positive (9/37; 24.3%), with more positive blood cultures in the FG (7/37, 18.9%) compared with the SG (2/37, 5.4%) (P = 0.03).
Intraoperative cultures were more frequently positive than blood cultures, with 24 out of 37 (64.8%) showing growth. A higher number of positive intraoperative cultures was found in the FG (14/37; 37.8%) compared with the SG (10/37; 27.0%) (P = 0.08).
The most isolated microorganism in both blood (4/9, 44.5%) and intraoperative (12/24, 50.0%) cultures was S. aureus. This microorganism was most frequently isolated in the FG compared with the SG in both blood (4/9, 44.5%) (0/9, 0.0%) (P = 0.44) and intraoperative (8/24, 33.3%) (4/24, 16.7%) (P = 0.32) cultures.
Previous reported infection and antibiotic treatment (Tables 3 and 4)
Table 3.
Relationship between previous reported infection and blood cultures with previous antibiotic treatment and DAIR outcome
| DAIR outcome | Total | P value | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Successful (n = 20) | Failure (n = 17) | ||||||||
| Culture | BC | BC ATB | BC | BC ATB |
BC | BC ATB | P1-value | P2- value | |
| RPI (n = 26), n (%) | Positive | 2 (5.4) | 2 (5.4) | 7 (18.9) | 3 (8.1) | 9 (24.3) | 5 (13.5) | 0.03* | 0.16 |
| Negative | 11 (29.7) | 7 (18.9) | 6 (16.2) | 2 (5.4) | 17 (45.9) | 9 (24.3) | |||
| NRPI (n = 11), n (%) | Positive | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | – | – |
| Negative | 7 (18.9) | 2 (5.4) | 4 (10.8) | 1 (2.7) | 11 (29.7) | 3 (8.1) | |||
| P-value3 | 0.27 | 0.46 | 0.64 | 0.10 | 0.27 | 0.22 | |||
Results are displayed as absolute value and percentage of the whole sample
RPI, reported previous infection; NRPI, non-reported previous infection; BC, blood culture; IC, intraoperative culture; ATB, previous antibiotic treatment; SG, successful group; FG, failure group
P1-value: statistical significance between the SG and FG in patients with positive or negative blood cultures who reported or did not report a previous infection
P2-value: statistical significance between the SF and FG in patients with positive or negative blood culture receiving antibiotic treatment, with reported or non-reported previous infection
P3-value: statistical significance between patients with reported previous infection (RPI) and those with non-reported previous infection (NRPI)
*P-value < 0.05
Table 4.
Relationship between previous reported infection and intraoperative cultures with previous antibiotic treatment and DAIR outcome
| DAIR outcome | Total | P value | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Successful (n = 20) | Failure (n = 17) | ||||||||
| Culture | IC | IC ATB | IC | IC ATB |
IC | IC ATB | P1-value | P2- value | |
| RPI (n = 26), n (%) | Positive | 7 (18.9) | 5 (13.5) | 11 (29.7) | 5 (13.5) | 18 (48.6) | 10 (27.0) | 0.09 | 0.08 |
| Negative | 6 (16.2) | 4 (10.8) | 2 (5.4) | 0 (0.0) | 8 (21.6) | 4 10.8) | |||
| NRPI (n = 11), n (%) | Positive | 3 (8.1) | 0 (0.0) | 3 (8.1) | 0 (0.0) | 6 (16.2) | 0 (0.0) | 0.30 | - |
| Negative | 4 (10.8) | 2 (5.4) | 1 (2.7) | 1 (2.7) | 5 (13.5) | 3 (8.1) | |||
| P-value3 | 0.64 | 0.45 | 0.66 | 0.16 | 0.40 | 0.05* | |||
Results are displayed as absolute value and percentage of the whole sample
RPI, reported previous infection; NRPI, non-reported previous infection; BC, blood culture; IC, intraoperative culture; ATB, previous antibiotic treatment; SG, successful group; FG, failure group
P1-value: statistical significance between the SG and FG in patients with positive or negative intraoperative cultures who reported or did not report a previous infection
P2-value: statistical significance between the SF and FG in patients with positive or negative intraoperative culture receiving antibiotic treatment, with reported or non-reported previous infection
P3-value: statistical significance between patients with reported previous infection (RPI) and those with non-reported previous infection (NRPI)
*P-value < 0.05
A total of 26 patients (70.3%) reported experiencing an infection episode prior to the onset of prosthetic pain. Among these patients, a higher percentage had positive blood cultures and positive intraoperative cultures compared with those without a history of prior infection (P = 0.03 and P = 0.40, respectively). Additionally, a higher proportion of patients who had received antibiotic treatment also had positive intraoperative cultures compared with those without a prior infection (P = 0.05). However, no significant differences were found in the SG and FG when comparing patients with or without a prior infection, in terms of positive or negative blood or intraoperative cultures, or previous antibiotic treatment.
In patients with a reported a previous infection, the failure rate was higher among those with positive blood cultures (p = 0.03). However, no significant differences in failure rates were observed in patients with positive blood cultures who had received antibiotic treatment (P = 0.16), those with positive intraoperative cultures (P = 0.09), or those with both positive intraoperative culture and prior antibiotic treatment (P = 0.08).
Among patients without a history of prior infection, no significant differences in failure rates were found based on positive or negative blood or intraoperative cultures, or the use of prior antibiotic treatment.
Source of hematogenous spread (Table 5)
Table 5.
Source of isolated microorganism in blood and intra-operative cultures
| Source HS | BC (N = 37) | IC (N = 37) | P1- value | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| SG | FG | SG | FG | |||||||
| P (n = 2) | N (n = 18) | P (n = 7) | N (n = 10) | P (n = 10) | N (n = 10) | P (n = 14) | N (n = 3) | |||
| RPI (n = 26) | Skin and soft tissues | 1 (2.7) | 4 (10.8) | 4 (10.8) | 2 (5.4) | 5 (13.5) | 0 (0.0) | 4 (10.8) | 1 (2.7) | 0.44 |
| Cardiovascular system | 0 (0.0) | 0 (0.0) | 2 (5.4) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 2 (5.4) | 0 (0.0) | ||
| Urogenital tract | 0 (0.0) | 4 (10.8) | 1 (2.7) | 3 (8.1) | 0 (0.0) | 4 (10.8) | 4 (10.8) | 0 (0.0) | ||
| Digestive tract | 1 (2.7) | 2 (5.4) | 0 (0.0) | 1 (2.7) | 1 (2.7) | 0 (0.0) | 1 (2.7) | 0 (0.0) | ||
| Respiratory tract | 0 (0.0) | 1 (2.7) | 0 (0.0) | 0 (0.0) | 1 (2.7) | 0 (0.0) | 0 (0.0) | 1 (2.7) | ||
| Dental origin | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 2 (5.4) | 0 (0.0) | 0 (0.0) | ||
| Total | 2 (5.4) | 11 (29.7) | 7 (18.9) | 6 (16.2) | 7 (18.9) | 6 (16.2) | 11 (29.7) | 2 (5.4) | ||
| NRPI (n = 11) | Skin and soft tissues | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 3 (8.1) | 0 (0.0) | 3 (8.1) | 0 (0.0) | 0.31 |
| Unknown | 0 (0.0) | 7 (18.9) | 0 (0.0) | 4 (10.8) | 0 (0.0) | 4 (10.8) | 0 (0.0) | 1 (2.7) | ||
| Total | 0 (0.0) | 7 (18.9) | 0 (0.0) | 4 (10.8) | 3 (8.1) | 4 (10.8) | 3 (8.1) | 1 (2.7) | ||
| P2-value | – | 0.84 | – | 0.62 | 0.60 | 0.85 | 0.50 | 0.87 | ||
Results are presented as absolute values and percentages of the whole sample
RPI, reported previous infection; NRPI, non-reported previous infection; HS, hematogenous spread; BC, blood culture; IC, intra-operative culture; SG, successful group; FG, failure group
P1- value: statistical significance between the source of hematogenous spread (HS) and the SG or FG outcome in patients with positive or negative blood or intraoperative cultures
P2 -value: statistical significance between patients with reported previous infection (RPI) and those with non-reported previous infection (NRPI)
*P-value < 0.05
In patients with positive blood and intraoperative cultures, the most common source of hematogenous spread was the skin and soft tissues, followed by the cardiovascular system, regardless of whether the patient had a prior history of infection. Other identified sources included the urogenital and digestive tracts.
No differences in failure proportion were observed on the basis of the source of hematogenous spread, nor whether blood or intraoperative cultures were positive or negative, in patients with a prior infection (p = 0.44) or in those without a prior infection (P = 0.31). Additionally, no significant differences were found between patients with a reported prior infection versus those without, when considering the source of hematogenous spread and the results of positive or negative blood or intraoperative cultures in both the SG and FG.
Surgical and antibiotic treatment (Table 6)
Table 6.
Surgical and antibiotic treatment
| Variable | DAIR outcome | P-value | ||
|---|---|---|---|---|
| Successful (n = 20) | Failure (n = 17) | |||
| Primary surgical indication (%) | Osteoarthritis | 85.0 | 100.0 | 0.10 |
| Neoplasia condition | 25.0 | 0.0 | ||
| Type of surgery (%) | Primary | 47.8 | 52.2 | 0.33 |
| Revision | 64.3 | 35.7 | ||
| Interval to surgery (days) | 6 ± 6.5 | 5.5 ± 7.5 | 0.83 | |
| Surgeon (%) | Expert | 80 | 76.5 | 0.80 |
| Nonexpert | 20 | 23.5 | ||
| PE exchange (%) | Yes | 100 | 80 | 0.07 |
| No | 0 | 20 | ||
| ATB duration (weeks) | 12 ± 4 | 8 ± 7 | 0.07 | |
Results displayed as percentages or absolute value
PE, polyethylene; ATB, antibiotic
*P-value < 0.05
No significant differences were observed between the SG and FG regarding the primary surgical indication for TKA (osteoarthritis versus other diagnosis) (P = 0.10), whether the surgery was a primary or revision TKA (P = 0.33), the interval between symptoms onset and surgery (P = 0.83), and the surgeon’s level of experience (P = 0.80). However, a trend toward a more favorable outcome was observed when polyethylene exchange was performed in all cases (P = 0.07) and when the duration of antibiotic treatment was longer (P = 0.07).
DAIR risk factors (Fig. 1)
Fig. 1.
Analysis of DAIR risk factors. OR, odds ratio; CI, confidence Interval; PE, polyethylene. *P-value < 0.05
Logistic regression analysis identified positive blood cultures as the only significant predictor of DAIR failure (OR 12, 95% CI: 1.1−18, P = 0.04), after adjusting for all confounding variables. The model demonstrated an overall accuracy of 80%, with a sensitivity of 73.0%, specificity of 85%, and an AUC of 89.2%. None of the other variables analyzed, viz. age, sex , comorbidities (rheumatoid arthritis and COPD), CRP serum levels, indication of TKA, positive intraoperative cultures, microorganism (S. aureus), or polyethylene exchange, were predictive of DAIR failure.
Then, a parsimonious Firth bias-reduced multivariable logistic regression model was fit, adjusting for host-related factors (CRIME80 variables), polyethylene exchange, and blood and intraoperative cultures. In this final model, blood culture positivity remained independently associated with DAIR failure (adjusted OR 4.5, 95% CI 1.02–23.6; p = 0.04). Failure to perform polyethylene exchange was also strongly associated with DAIR failure, although estimates remained imprecise owing to the small number of cases.
Discussion
The most important finding of this study was that the presence of positive blood cultures in our cohort of patients with late acute hematogenous total knee arthroplasty (LAH TKA) infection significantly increased the risk of DAIR failure. Other findings include that positive blood cultures were more common in patients with a history of prior infection. Although not statistically significant, positive intraoperative cultures were more frequent in patients for whom DAIR treatment failed, particularly in those with a prior history of infection, even if they were receiving antibiotic treatment. A trend toward increased DAIR failure was observed in patients with higher serum PCR levels, no polyethylene exchange during DAIR surgery, and a shorter duration of post-DAIR antibiotic treatment.
Hematogenous PJI occurs months or even years after TKA, resulting from the spread of bacteria through the bloodstream to a previously well-functioning prosthesis [5, 6]. These infections represent approximately 20% of all PJIs, though prevalence rates vary across studies [7, 8]. In most cases, patients report a prior infection episode that is related to their current condition [17, 18]. In our study, 26 (70.3%) out of 37 patients reported a previous infection before their current presentation, highlighting the importance of blood cultures in suspected hematogenous spread cases [19]. However, blood cultures are not always performed, depending on factors such as surgeon judgment or the presence of fever [14, 15]. In our cohort, although 50 patients were diagnosed with presumed hematogenous PJI, blood cultures were obtained in only 37 cases. The diagnosis of hematogenous infection is often challenging, as patients may not recall preceding infectious episodes, blood cultures may not be requested at admission, or acute presentations may represent exacerbations of chronic PJI rather than true hematogenous events [8, 11]. To ensure accurate classification, TKA infections were considered truly hematogenous only when they fulfilled the Delphi International Consensus diagnostic criteria [16] and blood cultures were requested at the time of admission.
Because the absence of blood cultures may reflect a lower clinical suspicion of true hematogenous infection, potentially indicating chronic PJI with acute exacerbation, we performed sensitivity analyses to explore how alternative assumptions regarding these patients could influence the estimated association between blood culture positivity and DAIR failure. The results of these analyses supported the robustness of the observed association and reinforced the clinical relevance of blood culture positivity as a marker of increased risk of DAIR failure in suspected hematogenous PJI.
The primary sources of bacteria in hematogenous TKA infections include the skin and soft tissues, cardiovascular system, oral cavity, and the gastrointestinal or urogenital tracts [7, 9, 18]. Our findings are consistent with this distribution, with Staphylococcus aureus being the most frequently isolated microorganism, in line with previous reports [7, 9, 18].
Not all TKA hematogenous infections yield positive blood cultures. Negative results can occur owing to: (1) prior antibiotic treatment; (2) low bacterial load or intermittent bacteremia; (3) errors in blood sample collection, processing or storage; or (4) misdiagnosis of the infection [14, 20, 21]. In our cohort, only nine patients had positive blood cultures, which were significantly more frequent among patients who reported a prior infection compared with those who did not (P = 0.03). Although antibiotic treatment could potentially mask bacteremia, we found no significant differences in blood culture results between patients who were receiving antibiotics at admission and those who were not. We hypothesized that bacterial burden, rather than prior antibiotic exposure, may be a key factor influencing blood culture positivity, even among patients with a reported prior infection. Notably, all 11 patients who did not report a prior infection had negative blood cultures, which may reflect a lower bacterial load or, alternatively, possible misclassification of the infection mechanism.
A higher proportion of patients had positive intraoperative cultures compared with positive blood cultures, suggesting that bacteremia in hematogenous TKA infection is often transient rather than persistent and may not always be captured by blood cultures. In our cohort, positive intraoperative cultures were more frequently observed among patients who experienced DAIR failure, particularly in the RPI subgroup, although this association did not reach statistical significance. This trend was observed both in patients who had received antibiotic therapy prior to surgery and in those who had not. These findings reinforce the importance of obtaining intraoperative cultures, as positive results may reflect a higher bacterial burden, whose impact on infection control may also depend on the virulence and characteristics of the isolated microorganism, despite adequate debridement and antimicrobial therapy [22, 23].
The treatment of LAH PJIs can involve debridement, antibiotics, and implant retention (DAIR) or revision surgery depending on factors such as duration of symptoms, implant stability (well-fixed or loose), or the type of microorganism [1, 10]. DAIR is a widely accepted treatment for hematogenous infections, especially when symptoms are recent and the implant is stable. The goal is to preserve a functional and stable implant [1, 10, 11]. DAIR should be performed as soon as the diagnosis is established, although the causative microorganism is rarely identified at that time. Success rates for DAIR in hematogenous PJI are consistently lower than those for acute infections [12, 24], with reported success rates ranging from 20% and 60%, depending on the series [12, 15]. The reasons of this variation in outcomes between hematogenous and acute infections remain unclear [8].
DAIR failure is defined as the inability to completely eradicate the infection, requiring additional interventions such as repeat DAIR, revision surgery, long-term antibiotic treatment, or infection-related mortality [16]. Positive blood cultures were more common in the FG (p = 0.03), being the only significant predictor of DAIR failure (OR 12, 95% CI: 1.1−18, p = 0.04; adjusted OR 4.48, 95% CI 1.02–23.6; p = 0.04). These findings suggest that obtaining blood cultures may be considered in patients with clinical suspicion of LAH TKA infection.
The impact of positive blood cultures on DAIR treatment and prognosis remains poorly understood, with limited studies investigating this relationship [14, 15]. Klement et al. found that positive blood cultures in PJI were associated with reduced treatment success. However, their study included all types of PJIs and various surgical treatments, which may have introduced confounding factors [14]. Kuo et al. focused exclusively on hematogenous infections and found that positive blood cultures were linked to poorer outcomes, but their study included both hip and knee prosthetic infections, which may respond differently to DAIR [15]. Anatomical and biomechanical differences between the hip and knee joints, and the knee’s greater structural complexity and range of motion, are thought to promote biofilm formation and may complicate debridement in knee infections [25]. Although our study is the only one focused especially on LAH TKA, which may lead to different results, our findings align with the poor outcomes described when positive blood cultures are present.
Several risk factors for treatment failure in periprosthetic knee infection have been identified across different surgical strategies, including both DAIR and two-stage revision approaches [26, 27]. In the context of DAIR, the CRIME80 scoring system has been proposed to quantify these risks [11, 28]. Notable risk factors include the causative microorganism, particularly S. aureus [28], as well as conditions such as rheumatoid arthritis, chronic obstructive pulmonary disease, CRP levels above 150 mg/L, male sex , age over 80 years, and the presence of a fracture that necessitated the prosthesis [11]. Shorter antibiotic treatment regimens (< 6 weeks) have also been linked to higher failure rates [10, 11], and the absence of polyethylene exchange further reduces DAIR success rate [10, 11]. However, the CRIME80 score may be more effective for hip prostheses than knee prostheses, given differences in indications [29] and cementation practices [30].
Although the risk factors considered in the CRIME80 score were prevalent in our cohort, we could not identify them as significant predictors of DAIR failure. However, higher serum CRP levels were observed in the failure group, consistent with findings in the CRIME80 score [10, 11]. Although CRP levels correlate with clinical severity, they did not differ significantly between outcome groups in our cohort, and were therefore considered a marker of disease severity rather than an independent predictor of DAIR failure. [31, 32]. As previously reported, failure rates are higher when polyethylene exchange is not performed during DAIR [10, 11], likely owing to the inability to adequately debride the posterior capsule [10]. In our cohort, three patients did not undergo polyethylene exchange because the procedures were performed under emergent conditions in which polyethylene components were unavailable. Notably, all three of these patients experienced treatment failure. Although the small number of cases does not allow firm conclusions to be drawn, the fact that all patients without polyethylene exchange experienced treatment failure is consistent with previous reports, highlighting the role of polyethylene exchange in infection control during DAIR [10, 11].
A minimum antibiotic treatment duration of 6 weeks is recommended for successful outcomes [10, 33, 34]. According to our institutional protocol, antibiotic treatment following DAIR is administered for a total duration ranging from 8 to 12 weeks, depending on clinical tolerance and follow-up laboratory results. Patients in the successful DAIR group tended to receive longer overall antibiotic regimens. In our institution, patients with positive blood cultures are routinely treated with a longer antibiotic course as part of a standardized protocol. However, in our cohort, the overall duration of antibiotic therapy did not differ significantly between patients with DAIR success and failure, suggesting that antibiotic treatment length itself was unlikely to represent an independent determinant of DAIR outcome.
Our study had several limitations. The relatively small sample size may have limited the precision of some estimates, resulting in wide confidence intervals, particularly in subgroup analyses with few events. These reflect sparse data rather than calculation errors, as appropriate exact methods were applied. Despite this, results were consistent across analyses and should be interpreted as hypothesis-generating. Second, owing to the retrospective design, blood cultures were not obtained in all patients with suspected late acute hematogenous TKA infection. However, the consistency of diagnostic and inclusion criteria, together with the sensitivity analyses performed, supports the robustness of the results. Third, the relatively short follow-up period (12 months) limits the assessment of longer-term outcomes, as reactivation of TKA infections may occur beyond the first postoperative year. Finally, several other factors, such as patient’s nutritional status, blood transfusions, and the type of TKA implant, may influence clinical outcome. These factors were not studied here and warrant further investigation.
Conclusions
Positive blood cultures in patients with late acute hematogenous TKA infection should be considered a risk factor for DAIR failure. Blood culture positivity was more frequent in patients with a history of prior infection, while positive intraoperative cultures were more commonly observed in patients who experienced DAIR failure, particularly among those with a previous infection history, even when antibiotic therapy had been initiated.
A trend toward increased DAIR failure was also observed in patients presenting with higher serum CRP levels, absence of polyethylene exchange during DAIR surgery, and a shorter duration of postoperative antibiotic treatment.
These findings suggest that patients with positive blood or intraoperative cultures may harbor a higher bacterial burden, resulting in a more pronounced systemic inflammatory response, as reflected by elevated serum CRP levels, and consequently poorer outcomes. This effect may be exacerbated when surgical (polyethylene exchange) and medical (adequate duration of antibiotic therapy) management are suboptimal.
Recognition of these associations is essential for optimizing treatment strategies, potentially supporting the use of more aggressive and prolonged antibiotic regimens and earlier, more comprehensive surgical intervention in selected patients.
Acknowledgements
Not applied.
Abbreviations
- LAH
Late acute hematogenous
- TKA
Total knee arthroplasty
- DAIR
Debridement, antibiotics, and implant retention
- BC
Blood culture
- PJI
Prosthetic joint infection
- COPD
Chronic obstructive pulmonary disease
- CRP
C-reactive protein
- BMI
Body mass index
- CKD
Chronic kidney disease
- PMN
Polymorphonuclear
- IV
Intravenous
- SD
Standard deviation
- IQR
Interquartile range
- RA
Rheumatoid arthritis
- HTN
Hypertension
- AUC
Area under the curve
- OD
Odds ratio
Author contributions
GPP participated in data acquisition, statistical analysis, interpretation of data, and drafting the article. ER, SM, JD, AV, EGO, VL, and RP participated in the provision of study patients, in technical and logistic support, and acquisition of data. JAHH participated in the conception and design of the study, analysis and interpretation of data, drafting the article, and final supervision.
Funding
This study had no funding.
Data availability
The datasets generated and/or analyzed during the current study are not publicly available owing to Institutional Board Policy but are available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
This study was approved by the institutional research ethics committee.
Competing interests
The authors declare that they have no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Gehrke T, Alijanipour P, Parvizi J (2015) The management of an infected total knee arthroplasty. Bone Joint J. 97-B(10 Suppl A):20–29 [DOI] [PubMed] [Google Scholar]
- 2.Shohat N, Bauer T, Buttaro M, Budhiparama N, Cashman J, Della Valle CJ, et al. (2019) Hip and knee section, what is the definition of a Periprosthetic Joint Infection (PJI) of the Knee and the Hip? Can the same criteria be used for both joints?: Proceedings of international consensus on orthopedic infections. J Arthroplast. 34(2): S325–7. [DOI] [PubMed]
- 3.Boddapati V, Fu MC, Mayman DJ, Su EP, Sculco PK, McLawhorn AS (2018) Revision total knee arthroplasty for periprosthetic joint infection is associated with increased postoperative morbidity and mortality relative to noninfectious revisions. J Arthroplasty 33(2):521–526 [DOI] [PubMed] [Google Scholar]
- 4.Sandiford NA, Francescini M, Kendoff D (2021) The burden of prosthetic joint infection (PJI). Ann Jt. 10.21037/aoj-2020-pji-11 [Google Scholar]
- 5.Zimmerli W, Trampuz A, Ochsner PE (2004) Prosthetic-joint infections. N Engl J Med. 10.1056/NEJMra040181 [DOI] [PubMed] [Google Scholar]
- 6.Tsukayama DT, Goldberg VM, Kyle R (2003) Diagnosis and management of infection after total knee arthroplasty. J Bone Joint Surg Am. 10.2106/00004623-200300001-00014 [DOI] [PubMed] [Google Scholar]
- 7.Rakow A, Perka C, Trampuz A, Renz N (2019) Origin and characteristics of haematogenous periprosthetic joint infection. Clin Microbiol Infect 25(7):845–850 [DOI] [PubMed] [Google Scholar]
- 8.Renz N, Trampuz A, Perka C, Rakow A (2022) Outcome and failure analysis of 132 episodes of hematogenous periprosthetic joint infections—a cohort study. Open Forum Infect Dis. 10.1093/ofid/ofac094 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zeller V, Kerroumi Y, Meyssonnier V, Heym B, Metten MA, Desplaces N et al (2018) Analysis of postoperative and hematogenous prosthetic joint-infection microbiological patterns in a large cohort. J Infect 76(4):328–334 [DOI] [PubMed] [Google Scholar]
- 10.Argenson JN, Arndt M, Babis G, Battenberg A, Budhiparama N, Catani F, et al. (2019) Hip and knee section, treatment, debridement and retention of implant: Proceedings of international consensus on orthopedic infections. J Arthroplast. 34(2): S399–419. [DOI] [PubMed]
- 11.Wouthuyzen-Bakker M, Sebillotte M, Lomas J, Taylor A, Palomares EB, Murillo O et al (2019) Clinical outcome and risk factors for failure in late acute prosthetic joint infections treated with debridement and implant retention. J Infect 78(1):40–47 [DOI] [PubMed] [Google Scholar]
- 12.Vilchez F, Martínez-Pastor JC, García-Ramiro S, Bori G, Tornero E, García E et al (2011) Efficacy of debridement in hematogenous and early post-surgical prosthetic joint infections. Int J Artif Organs 34(9):863–869 [DOI] [PubMed] [Google Scholar]
- 13.Iza K, Foruria X, Moreta J, Uriarte I, Loroño A, Aguirre U et al (2019) DAIR (debridement, antibiotics and implant retention) less effective in hematogenous total knee arthroplasty infections. J Orthop Surg Res 14(1):1–6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Klement MR, Siddiqi A, Rock JM, Chen AF, Bolognesi MP, Seyler TM (2018) Positive blood cultures in periprosthetic joint infection decrease rate of treatment success. J Arthroplasty 33(1):200-204.e1 [DOI] [PubMed] [Google Scholar]
- 15.Kuo FC, Goswami K, Klement MR, Shohat N, Parvizi J (2019) Positive blood cultures decrease the treatment success in acute hematogenous periprosthetic joint infection treated with debridement, antibiotics, and implant retention. J Arthroplasty 34(12):3030-3034.e1 [DOI] [PubMed] [Google Scholar]
- 16.Diaz-Ledezma C, Higuera CA, Parvizi J (2013) Success after treatment of periprosthetic joint infection: a delphi-based international multidisciplinary consensus. Clin Orthop Relat Res 471(7):2374 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Swan J, Dowsey M, Babazadeh S, Mandaleson A, Choong PFM (2011) Significance of sentinel infective events in haematogenous prosthetic knee infections. ANZ J Surg 81(1–2):40–45 [DOI] [PubMed] [Google Scholar]
- 18.Honkanen M, Jämsen E, Karppelin M, Huttunen R, Eskelinen A, Syrjänen J (2019) Periprosthetic joint infections as a consequence of bacteremia. Open Forum Infect Dis. 10.1093/ofid/ofz218 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Sconfienza LM, Signore A, Cassar-Pullicino V, Cataldo MA, Gheysens O, Borens O et al (2019) Diagnosis of peripheral bone and prosthetic joint infections: overview on the consensus documents by the EANM, EBJIS, and ESR (with ESCMID endorsement). Eur Radiol 29(12):6425–6438 [DOI] [PubMed] [Google Scholar]
- 20.Kalbian I, Park JW, Goswami K, Lee YK, Parvizi J, Koo KH (2020) Culture-negative periprosthetic joint infection: prevalence, aetiology, evaluation, recommendations, and treatment. Int Orthop 44(7):1255–1261 [DOI] [PubMed] [Google Scholar]
- 21.Seifert H (2009) The Clinical importance of microbiological findings in the diagnosis and management of bloodstream infections. Clin Infect Dis 48(Suppl 4):S238–S245 [DOI] [PubMed] [Google Scholar]
- 22.Tande AJ, Patel R (2014) Prosthetic joint infection. Clin Microbiol Rev 27(2):302 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Zendeli F, Jędrusik A, Schaefer RO, Albrecht D, Betz M, Waibel FWA et al (2025) Pathogen-specific risk for iterative surgical debridement in orthopedic infections: a prospective multicohort analysis. J Clin Med 14(24):8750 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Knoll L, Steppacher SD, Furrer H, Thurnheer-Zürcher MC, Renz N (2023) High treatment failure rate in haematogenous compared to non-haematogenous periprosthetic joint infection. Bone Joint J 105-B(12):1294–1302 [DOI] [PubMed] [Google Scholar]
- 25.Scheper H, Gerritsen LM, Pijls BG, Van Asten SA, Visser LG, De Boer MGJ (2021) Outcome of debridement, antibiotics, and implant retention for staphylococcal hip and knee prosthetic joint infections, focused on rifampicin use: a systematic review and meta-analysis. Open Forum Infect Dis. 10.1093/ofid/ofab298 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Puetzler J, Hofschneider M, Gosheger G, Theil C, Schulze M, Schwarze J et al (2024) Evaluation of time to reimplantation as a risk factor in two-stage revision with static spacers for periprosthetic knee joint infection. J Orthop Traumatol. 10.1186/s10195-024-00745-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lee WS, Park KK, Cho BW, Park JY, Kim I, Kwon HM (2024) Risk factors for early septic failure after two-stage exchange total knee arthroplasty for treatment of periprosthetic joint infection. J Orthop Traumatol. 10.1186/s10195-024-00750-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Wouthuyzen-Bakker M, Sebillotte M, Lomas J, Kendrick B, Palomares EB, Murillo O et al (2019) Timing of implant-removal in late acute periprosthetic joint infection: a multicenter observational study. J Infect 79(3):199–205 [DOI] [PubMed] [Google Scholar]
- 29.Qin CD, Helfrich MM, Fitz DW, Hardt KD, Beal MD, Manning DW (2017) The Lawrence D. Dorr Surgical Techniques & technologies award: differences in postoperative outcomes between total hip arthroplasty for fracture vs osteoarthritis. J Arthroplast. 32(9S):S3-7 [DOI] [PubMed] [Google Scholar]
- 30.Prasad AK, Tan JHS, Hanna SA, Dawson-Bowling S, Bedair HS (2020) Cemented vs. cementless fixation in primary total knee arthroplasty: a systematic review and meta-analysis. EFORT Open Rev 5(11):793–798 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Ludwick L, Siqueira M, Shohat N, Sherman MB, Streicher S, Parvizi J (2022) For patients with acute PJI treated with debridement, antibiotics, and implant retention, what factors are associated with systemic sepsis and recurrent or persistent infection in septic patients? Clin Orthop Relat Res 480(8):1491–1500 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Kheir MM, Tan TL, Shohat N, Foltz C, Parvizi J (2018) Routine diagnostic tests for periprosthetic joint infection demonstrate a high false-negative rate and are influenced by the infecting organism. J Bone Joint Surg 100(23):2057–2065 [DOI] [PubMed] [Google Scholar]
- 33.Lora-Tamayo J, Mancheño-Losa M, Meléndez-Carmona MÁ, Hernández-Jiménez P, Benito N, Murillo O (2024) Appropriate duration of antimicrobial treatment for prosthetic joint infections: a narrative review. Antibiotics. 10.3390/antibiotics13040293 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Miller R, Higuera CA, Wu J, Klika A, Babic M, Piuzzi NS (2020) Periprosthetic joint infection: a review of antibiotic treatment. JBJS Rev. 10.2106/JBJS.RVW.19.00224 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated and/or analyzed during the current study are not publicly available owing to Institutional Board Policy but are available from the corresponding author on reasonable request.