If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Requests for reprints should be addressed to Aaron J. Tande, MD, Division of Infectious Diseases, Department of Internal Medicine, Mayo Clinic, 200 1st Street SW, Rochester, MN 55905.
Staphylococcus aureus bacteremia is a life-threatening condition that may lead to metastatic infection, including prosthetic joint infection.
Methods
To assess clinical factors associated with hematogenous prosthetic joint infection, we retrospectively reviewed all patients with a joint arthroplasty in place at the time of a first episode of S. aureus bacteremia over a 5-year period at our institution. Patients with postsurgical prosthetic joint infection without hematogenous prosthetic joint infection were excluded.
Results
There were 85 patients (143 arthroplasties) with either no prosthetic joint infection (n = 50; 58.8%) or hematogenous prosthetic joint infection in at least one arthroplasty (n = 35; 41.2%). The odds of hematogenous prosthetic joint infection was significantly increased among patients with community-acquired S. aureus bacteremia (odds ratio [OR] 18.07; 95% confidence interval [CI] 2.64-infinity; P = .001), as compared with nosocomial S. aureus bacteremia, in which there were no patients with hematogenous prosthetic joint infection. After adjusting for S. aureus bacteremia classification, the presence of ≥3 joint arthroplasties in place was associated with a nearly ninefold increased odds of hematogenous prosthetic joint infection as compared with those with 1-2 joint arthroplasties in place (OR 8.55; 95% CI 1.44-95.71; P = .012). All but one joint with prosthetic joint infection demonstrated at least one clinical feature suggestive of infection. There were 4 additional S. aureus prosthetic joint infections diagnosed during a median of 3.4 years of follow-up post hospitalization for S. aureus bacteremia.
Conclusion
Prosthetic joint infection is frequent in patients with existing arthroplasties and concomitant S. aureus bacteremia, particularly with community-acquired S. aureus bacteremia and multiple prostheses. In contrast, occult S. aureus prosthetic joint infection without clinical features suggestive of prosthetic joint infection at the time of S. aureus bacteremia is rare.
Community-acquired Staphylococcus aureus bacteremia and 3 or more arthroplasties are associated with increased risk of prosthetic joint infection during S. aureus bacteremia.
•
Joint-specific associations with prosthetic joint infection include knee arthroplasty and prior revision surgery.
•
While active investigation for prosthetic joint infection should be pursued in any symptomatic patient, occult prosthetic joint infection is rare.
Introduction
Staphylococcus aureus bacteremia is the second leading cause of nosocomial-onset
Complications of S. aureus bacteremia that are unrecognized at the time of S. aureus bacteremia diagnosis may lead to inadequate therapy, increased morbidity, and relapse of infection.
Joint replacement is one of the most successful and frequently performed medical procedures. Over 1 million total hip and knee arthroplasties were performed in 2010 in the United States,
joint arthroplasties remain vulnerable to hematogenous seeding throughout the lifespan of the prosthesis.
Several smaller studies have suggested that prosthetic joint infection occurs in 30% to 40% of patients with joint prostheses in place at the time of S. aureus bacteremia.
These studies suggested trends toward increased risk of prosthetic joint infection among patients with diabetes mellitus, community-acquired S. aureus bacteremia, knee arthroplasty, and methicillin-susceptible S. aureus bacteremia, but did not provide detailed analyses of the clinical presentations and orthopedic characteristics of patients with and without prosthetic joint infection. The purpose of this study was to identify clinical predictors of prosthetic joint infection in patients presenting with S. aureus bacteremia.
Methods
Study Setting and Participants
Patients hospitalized from June 1, 2006 to June 30, 2011 with S. aureus bacteremia at our institution were included in this analysis. Methodology for identification of S. aureus bacteremia cases, and inclusion and exclusion criteria are described in an earlier publication from this S. aureus bacteremia cohort.
Predicting Risk of Endocarditis Using a Clinical Tool (PREDICT): scoring system to guide use of echocardiography in the management of Staphylococcus aureus bacteremia.
The medical records of all patients in this database were reviewed, and all adults with a knee, hip, shoulder, or elbow arthroplasty in place at the time of their first episode of S. aureus bacteremia at our institution were included in this analysis. All patients provided consent to participate in research studies at Mayo Clinic. The study was approved by the Mayo Clinic Institutional Review Board.
Data Collection
Clinical data were obtained by review of the electronic medical records for all patients by one of the investigators (AJT or BRP). The definitions used are described in Table 1.
Cases in which the classification of hematogenous prosthetic joint infection vs primary/indeterminate prosthetic joint infection was not clear were reviewed with an additional author (DRO).
Table 1Definitions of Terms Used in this Study
Term
Definition
Definition of SAB
More than one peripheral blood culture positive for Staphylococcus aureus or one positive peripheral blood culture with signs and symptoms of infection.
SAB present or incubating at the time of admission to the hospital, or associated with the first positive blood culture obtained within 48 h of admission, in a patient who does not meet criteria for health care-associated infection.
Nosocomial SAB
SAB developing in a patient hospitalized for more than 48 h before the onset of signs/symptoms consistent with bacteremia.
Health care-associated, community-onset SAB
SAB diagnosed within 48 h of admission in an outpatient with any of the following criteria:
(1)
Received intravenous therapy, wound care, or specialized nursing care at home within the 30 d before the onset of SAB; or
(2)
Attended a hospital or hemodialysis clinic or received intravenous chemotherapy within the 30 d before the onset of SAB; or
(3)
Was hospitalized in an acute care hospital for 2 or more days in the 90 d before the onset of SAB; or
(4)
Resided in a nursing home or long-term care facility.
Duration of bacteremia
Based on the duration between date of first positive blood culture and first negative blood culture.
Immunocompromising condition
Active malignancy.
HIV.
Neutropenia (absolute neutrophil count <500 cells/μL at any time during SAB).
Prior hematopoietic stem cell or solid organ transplant.
Immunosuppressive therapy in the 3 months befpre SAB:
Systemic corticosteroids (more than 30 d),
Nonbiologic disease-modifying agents (cyclosporine, methotrexate, or azathioprine),
S. aureus isolated from one sterile synovial fluid or tissue culture with any of the following criteria:
(1)
S. aureus isolated from a second sterile synovial fluid or tissue culture; or
(2)
Sinus tract communicating with the prosthesis; or
(3)
4 of: Elevated ESR/CRP, synovial fluid WBC, synovial fluid PMN%, joint purulence, or acute inflammation on periprosthetic histopathologic examination; or
(4)
Two-investigator review.
Classification of PJI
Primary PJI
Clinical course suggesting direct inoculation from surgery or contiguous spread of infection.
Hematogenous PJI
Presence of an alternate source for the SAB or acute onset of symptoms of PJI in a previously asymptomatic joint not meeting definition of primary PJI.
Indeterminate PJI
PJI that could not be confidently classified as either hematogenous or primary PJI.
PJI treatment failure
(1)
Death related to SAB or PJI; or
(2)
Additional debridement for PJI >30 d or resection arthroplasty <2 y after a DAIR strategy; or
(3)
Resection for any reason <2 y after reimplantation among patients undergoing 2-stage exchange; or
(4)
Nonreimplantation due to ongoing PJI or subsequent surgery for PJI after reimplantation among subjects treated with resection.
CRP = C-reactive protein; DAIR = debridement and implant retention; ESR = erythrocyte sedimentation rate; HIV = human immunodeficiency virus; PJI = prosthetic joint infection; PMN% = neutrophil percentage; SAB = Staphylococcus aureus bacteremia; WBC = white blood cell count.
Following S. aureus bacteremia, medical records were reviewed until the latest adequate clinical visit or death in order to assess for S. aureus prosthetic joint infection not diagnosed during the initial hospitalization. Duration of follow-up was the difference between first positive blood culture and last recorded follow-up. Visits with an orthopedic surgeon or infectious diseases provider, a full physical examination with a general or subspecialty medical provider, or mail-in questionnaires obtained as part of the Mayo Clinic Joint Arthroplasty Registry
Correlation of patient questionnaire responses and physician history in grading clinical outcome following hip and knee arthroplasty. A prospective study of 201 joint arthroplasties.
were considered sufficient for follow-up information.
Statistical Analysis
Data were collected and entered into a secure REDCap Database (Vanderbilt University, Nashville, TN). Continuous features were summarized with medians and interquartile ranges (IQRs); categorical features were summarized with counts and percentages. Associations with hematogenous prosthetic joint infection among patients with S. aureus bacteremia and at least one joint arthroplasty were evaluated using logistic regression models. A multivariable model was developed using stepwise selection, with the P-value for a feature to enter or leave the model set to .05. Overall survival rates were estimated using the Kaplan-Meier method and compared between patients with and without hematogenous prosthetic joint infection using log-rank tests. The duration of follow-up for the survival analyses was calculated from the date of first diagnosis of S. aureus bacteremia to the date of death or last follow-up. Statistical analyses were performed using the SAS software package (SAS Institute, Inc, Cary, NC). All tests were 2-sided, and P-values <.05 were considered statistically significant.
Results
Of the 678 patients with S. aureus bacteremia in the study period, 97 (14.3%) patients had 166 arthroplasties in place at the time of bacteremia and were included in the present study. Fifty patients (51.6%) had no prosthetic joint infection, 35 (36.1%) had hematogenous prosthetic joint infection in at least one joint arthroplasty, and 12 (12.4%) had only primary postsurgical or indeterminate prosthetic joint infection, without hematogenous prosthetic joint infection. The 12 patients with only primary or indeterminate prosthetic joint infection were excluded, and the remaining 85 patients (143 arthroplasties) who had either no prosthetic joint infection (n = 50) or hematogenous prosthetic joint infection in at least one arthroplasty (n = 35), were included for analysis (Figure 1).
Figure 1Flowchart of the study cohort. PJI = prosthetic joint infection.
After treatment of a cellulitis surrounding a toe ulcer, this patient developed pain in the contralateral knee arthroplasty and presented with S. aureus bacteremia and signs of prosthetic joint infection. This patient underwent debridement, from which a single operative specimen was sent. This sample grew S. aureus with identical susceptibility results to the blood culture isolate. No preoperative aspiration was performed. Operative histopathology demonstrated acute inflammation, and the patient had a high erythrocyte sedimentation rate and C-reactive protein, but lacked enough supporting criteria to fulfill the MSIS criteria for prosthetic joint infection. This was felt to be consistent with prosthetic joint infection based on 2-investigator review.
Clinical Presentation at S. aureus Bacteremia Diagnosis
includes the comparative analysis among patients with hematogenous prosthetic joint infection and those with no prosthetic joint infection. There were several nonstatistically significant differences among patients with hematogenous prosthetic joint infection, as compared with those without prosthetic joint infection. There was no difference in the presence of endocarditis between the 2 groups. Other observed complications of S. aureus bacteremia included vertebral osteomyelitis/diskitis (n = 9) and cardiac implantable electronic device infection (n = 7). Seven patients had a central venous catheter in place at the time of S. aureus bacteremia, none of whom developed a hematogenous prosthetic joint infection (P = .041). The catheter was the likely source of S. aureus bacteremia in 6 of 7 patients and was removed in all 7 patients. The classification of S. aureus bacteremia appeared to correlate with diagnosis of hematogenous prosthetic joint infection, as there were no patients with nosocomial S. aureus bacteremia diagnosed with hematogenous prosthetic joint infection. In contrast, patients with community-acquired S. aureus bacteremia had an 18-fold increased odds of being diagnosed with hematogenous prosthetic joint infection, compared with those with nosocomial S. aureus bacteremia (odds ratio [OR] 18.07; 95% confidence interval [CI], 2.64-infinity; P = .001). Community-onset health care-associated S. aureus bacteremia was associated with nonstatistically significant increased odds of hematogenous prosthetic joint infection, compared with nosocomial S. aureus bacteremia (OR 6.15; 95% CI, 0.86-infinity; P = .075). The number of prostheses in situ also correlated with hematogenous prosthetic joint infection, as the presence of 3 or more arthroplasties was associated with a greater than fivefold increased odds of hematogenous prosthetic joint infection, compared with 2 or fewer arthroplasties (OR 5.42; 95% CI, 1.35-21.81; P = .017). Multivariable logistic regression analysis demonstrated that both the classification of S. aureus bacteremia and the number of joint arthroplasties in place were jointly significantly associated with hematogenous prosthetic joint infection (Table 3). After adjusting for the classification of S. aureus bacteremia, the presence of 3 or more joint arthroplasties in place was associated with nearly ninefold increased odds of hematogenous prosthetic joint infection, compared with patients with only 1-2 joint arthroplasties in place (OR 8.55; 95% CI, 1.44-95.71; P = .012).
Table 2Univariable Associations with Hematogenous PJI
Immunocompromising conditions included immunosuppressive medications (n = 22), active malignancy (n = 14), or neutropenia (n = 2). Patients could have more than one cause for immunocompromise.
Definite cases using the modified Duke Criteria.16
8 (16%)
3 (8.6%)
.32
Classification of SAB
Nosocomial
10 (20%)
0
Community acquired
17 (34%)
24 (68.6%)
.001
P-values were calculated through logistic regression modeling. In the presence of a 0 count, odds ratios were estimated using exact logistic regression.
Community onset health-care associated
23 (46%)
11 (31.4%)
.075
P-values were calculated through logistic regression modeling. In the presence of a 0 count, odds ratios were estimated using exact logistic regression.
Oxacillin susceptible
32 (64%)
29 (82.9%)
.062
Categorical variables expressed in absolute number and (percentage); continuous variables expressed in median and (interquartile range).
‡ Immunocompromising conditions included immunosuppressive medications (n = 22), active malignancy (n = 14), or neutropenia (n = 2). Patients could have more than one cause for immunocompromise.
§ Includes vascular or decubitus ulcers, chronic dermatitis, or psoriasis.
‖ ESR was known in 56 patients.
¶ CRP was known in 58 patients.
∗∗ Definite cases using the modified Duke Criteria.
†† P-values were calculated through logistic regression modeling. In the presence of a 0 count, odds ratios were estimated using exact logistic regression.
Table 3Associations with Hematogenous PJI on Multivariable Analysis
Characteristic
Odds Ratio (95% CI)
P-Value
Classification of SAB
Nosocomial
1.0 (reference)
Community onset health-care associated
5.91 (0.77-infinity)
.095
Community acquired
21.39 (2.92-infinity)
.001
Number of arthroplasties in place
1-2
1.0 (reference)
3 or more
8.55 (1.44-95.71)
.012
P-values were calculated through logistic regression modeling. In the presence of a zero count, odds ratios were estimated using exact logistic regression.
CI = confidence interval; PJI = prosthetic joint infection; SAB = Staphylococcus aureus bacteremia.
Overall, 85 patients had 143 arthroplasties in place: 47 patients had 1, 26 patients had 2, 7 patients had 3, 3 patients had 4, 1 patient had 5, and 1 patient had 6 arthroplasties. The arthroplasty locations were knee (n = 73), hip (n = 59), shoulder (n = 10), and elbow (n = 1). Among the 35 patients with at least one hematogenous prosthetic joint infection, there were 39 arthroplasties that were hematogenously infected, one arthroplasty that was a primary prosthetic joint infection, and 3 others that were indeterminate (Figure 1). After excluding the 4 primary or indeterminate prosthetic joint infections, there were 139 joint arthroplasties in 85 patients (Table 4). Previously revised joints were significantly more likely to develop hematogenous prosthetic joint infection (OR 3.35; 95% CI, 1.36-8.23; P = .01). When analysis was limited to only hip and knee arthroplasties, hematogenous prosthetic joint infection occurred in 25 (35.2%) of 71 knee arthroplasties and 11 (18.6%) of 59 hip arthroplasties (P = .036). Of the 8 shoulder arthroplasties, 3 (37.5%) developed hematogenous prosthetic joint infection. There was one elbow arthroplasty that did not develop infection.
Table 4Orthopedic Characteristics of the Joint Arthroplasties in Place at Time of SAB
The most common signs and symptoms of prosthetic joint infection were joint pain (97.4%), periarticular swelling or effusion (61.5%), and periarticular warmth (46.2%). In one patient, there was bilateral knee prosthetic joint infection, although there were no symptoms in one knee. This patient presented with S. aureus bacteremia secondary to postthoracotomy S. aureus pleural empyema and a painful right knee arthroplasty. Orthopedic surgery evaluation noted effusion and warmth about the right knee, but there were no abnormalities in the left hip or left knee arthroplasty. While performing debridement of the right knee, aspiration of the asymptomatic left knee confirmed the presence of S. aureus prosthetic joint infection.
Treatment of S. aureus Bacteremia
All patients received intravenous antimicrobial therapy for S. aureus bacteremia (Table 5). Nine (18%) patients without prosthetic joint infection and 4 (11.4%) patients with prosthetic joint infection died before completion of parenteral antimicrobial therapy. Among patients who completed parenteral antimicrobial therapy, the median duration of intravenous antimicrobials was >4 weeks in both groups. Among the 35 patients with hematogenous prosthetic joint infection, 32 underwent either resection of the arthroplasty, or irrigation and debridement with implant retention (DAIR) for treatment of prosthetic joint infection. The remaining 3 patients had significant medical comorbidities and were managed with comfort-focused care rather than surgical treatment; all 3 died within 60 days. Among those with prosthetic joint infection who underwent DAIR (n = 17), 15 were treated with a rifampin-based regimen, and subsequent chronic oral antimicrobial suppression was given to 17 patients. Two patients without prosthetic joint infection at presentation received rifampin.
Table 5Antimicrobial Therapy Used
No PJI (n = 50)
Hematogenous PJI (n = 35)
Subject died during intravenous antimicrobial course
9 (18%)
4 (11.4%)
Median duration of intravenous antimicrobial therapy, in d (IQR)
Rifampin was used in combination with DAIR procedure in patients with PJI and due to persistent bloodstream infection with multifocal soft tissue abscesses in the patient without.
1 (2%)
15 (42.9%)
Gentamicin and rifampin
1 (2%)
0
Oral antimicrobial(s) after intravenous antimicrobial
Reasons for subsequent oral antimicrobials include DAIR procedure (n = 17), spine infection (n = 7), concern for relapse after arthroplasty reimplantation (n = 3), retained spacer (n = 1), high risk of cardiac device seeding (n = 1), and vascular graft infection (n = 1).
8 (16%)
22 (62.9%)
Categorical variables expressed in absolute number and (percentage).
DAIR = debridement and implant retention; IQR = interquartile range; PJI = prosthetic joint infection.
∗ Among the 72 patients who survived to completion of intravenous antimicrobials.
† The definitive antimicrobial was the antimicrobial(s) used for ≥50% of the treatment course.
‡ Rifampin was used in combination with DAIR procedure in patients with PJI and due to persistent bloodstream infection with multifocal soft tissue abscesses in the patient without.
§ Reasons for subsequent oral antimicrobials include DAIR procedure (n = 17), spine infection (n = 7), concern for relapse after arthroplasty reimplantation (n = 3), retained spacer (n = 1), high risk of cardiac device seeding (n = 1), and vascular graft infection (n = 1).
Follow-up and Subsequent Prosthetic Joint Infection
At 30 days after the first positive blood culture, 32 (91.4%) of 35 and 42 (84%) of 50 patients with and without hematogenous prosthetic joint infection were alive, respectively (P = .51). During the follow-up period, 14 of 35 patients with hematogenous prosthetic joint infection and 29 of 50 patients without prosthetic joint infection died at a median of 11 and 4 months after S. aureus bacteremia diagnosis, respectively (Figure 2; P = .087). Among the surviving patients, the latest follow-up was a median of 3.3 years (IQR 2.6-5.3 years) and 3.4 years (IQR 1.4-4.5 years) after S. aureus bacteremia in those with and without hematogenous prosthetic joint infection, respectively. Prosthetic joint infection treatment was successful in 26 (74.3%) of 35 patients and 29 (67.4%) of 43 arthroplasties with prosthetic joint infection. Among the 9 patients with 14 prosthetic joint infections who failed treatment, reasons for treatment failure included arthroplasty resection/component resection after DAIR (n = 7 arthroplasties), death due to S. aureus bacteremia (n = 6 arthroplasties, 4 patients), and repeat debridement >30 days after DAIR (n = 1 arthroplasty). One patient with bilateral knee prosthetic joint infection was treated with bilateral DAIR, with successful treatment of the left knee and treatment failure of the right knee, despite ongoing suppressive antimicrobials. This patient experienced ongoing culture-positive right knee prosthetic joint infection and was treated with a one-stage exchange procedure 169 days after his S. aureus bacteremia episode.
Figure 2Overall survival among patients with and without hematogenous prosthetic joint infection. There was not a statistically significant difference in likelihood of mortality among patients with or without hematogenous prosthetic joint infection (P = .087). Dotted line = subjects with hematogenous prosthetic joint infection (n = 35); solid line = subjects without hematogenous prosthetic joint infection (n = 50). The rows below the Figure show the number of patients at risk of mortality at each year. PJI = prosthetic joint infection; SAB = Staphylococcus aureus bacteremia.
There were 4 patients who had S. aureus prosthetic joint infection during the follow-up period, all occurring in the 50 patients without prosthetic joint infection during the initial hospitalization; these 4 had complicated S. aureus bacteremia at initial hospitalization (Table 6). Patient 4 is classified as a possible S. aureus prosthetic joint infection. Nearly 18 months after being treated for S. aureus bacteremia with multiple metastatic foci, this patient developed increasing left knee arthroplasty pain. Knee aspiration, performed while on suppressive trimethoprim/sulfamethoxazole, revealed >4500 nucleated cells with negative cultures. Repeat aspiration following antimicrobial discontinuation had a similar cell count and negative cultures. He underwent arthroplasty resection, and operative inspection was consistent with prosthetic joint infection. However, cultures were negative and no histopathology was obtained. The treating clinician felt that S. aureus prosthetic joint infection was likely.
Table 6Details of the 4 Patients Diagnosed with Subsequent Staphylococcus aureus PJI During Follow-up Period
Patient
Patient Age at SAB, in y
Duration From Last Joint Surgery to SAB, in y
Foci of Initial SAB
Classification of SAB
Duration of Bacteremia, in d
Intravenous Antimicrobial Treatment Duration for SAB, in d
Suppressive Oral Antimicrobials After Intravenous Antimicrobials
Duration Between SAB Clearance and PJI Diagnosis, in d
HCA = health care associated; IE = infective endocarditis; MSSA = methicillin susceptible S. aureus; MRSA = methicillin resistant S. aureus; PJI = prosthetic joint infection; SAB = Staphylococcus aureus bacteremia; THA = total hip arthroplasty; TKA = total knee arthroplasty TMP/SMX = trimethoprim/sulfamethoxazole; VISA = Vancomycin intermediate S. aureus.
This is the largest cohort to date to address hematogenous infection of prosthetic joints in patients with S. aureus bacteremia, and the first to analyze risk factors associated with prosthetic joint infection in S. aureus bacteremia patients using multivariable analysis. After excluding patients who had only primary prosthetic joint infection, we observed hematogenous prosthetic joint infection in 41% of patients and 27% of arthroplasties. This frequency is similar to that of previous smaller studies
All of the hematogenous prosthetic joint infection cases in our patient cohort occurred with community-acquired or community-onset health care-associated S. aureus bacteremia, which supports existing literature that complicated S. aureus bacteremia is significantly more likely in these settings than in nosocomial S. aureus bacteremia.
Correspondingly, there were no cases of nosocomial S. aureus bacteremia or catheter-associated S. aureus bacteremia in which hematogenous prosthetic joint infection was diagnosed. This may have been due to rapid diagnosis and initiation of empiric antimicrobial therapy in nosocomial S. aureus bacteremia cases. This observation is consistent with an earlier smaller study where no episodes of hematogenous prosthetic joint infection were observed in nosocomial S. aureus bacteremia cases.
The presence of 3 or more arthroplasties remained significantly associated with hematogenous prosthetic joint infection, even after controlling for S. aureus bacteremia classification, a finding not previously observed. Similar to prior studies, we did not observe a significant association between hematogenous prosthetic joint infection and immunocompromising conditions.
Hematogenous prosthetic joint infection was significantly more frequent in knee arthroplasty as compared with hip arthroplasty, occurring nearly twice as often. These findings expand on prior reports showing nonsignificant trends of increased hematogenous prosthetic joint infection in knee arthroplasty during S. aureus bacteremia
perhaps owing to the more complex mechanics of the knee joint as compared with the hip. Previous arthroplasty revision was also associated with an increased risk of hematogenous prosthetic joint infection, a novel finding. One postulated mechanism for this may be the increased prosthesis size sometimes required for revision arthroplasty to compensate for previous bone loss. This may create more surface area for bacterial attachment during S. aureus bacteremia. There was no association between risk of hematogenous prosthetic joint infection and time from primary joint arthroplasty or prior surgery, similar to prior findings.
A finding of this study that is highly clinically relevant and not previously reported, is that 97% of hematogenously infected arthroplasties had at least one sign or symptom suggestive of infection in the affected joint. While no single sign or symptom is present in all cases of prosthetic joint infection, pain is present in more than 75% of cases of prosthetic joint infection in other studies.
Similarly, all but one episode of prosthetic joint infection was associated with pain in this study. While a sinus tract communicating with the arthroplasty is considered definitive evidence of prosthetic joint infection,
this was not seen here, supporting the acute hematogenous nature of these prosthetic joint infections. Among the 100 arthroplasties that did not display signs or symptoms suggestive of prosthetic joint infection at the time of S. aureus bacteremia, treated with a median of 32 days of intravenous antimicrobials, 96 remained free of prosthetic joint infection until death or loss to follow-up. Although we believe that antimicrobial therapy alone has a low likelihood of success in treatment of unrecognized prosthetic joint infection,
we cannot exclude the possibility that the treatment provided for S. aureus bacteremia was sufficient to prevent subsequent overt prosthetic joint infection. These data suggest that while prosthetic joint infection associated with S. aureus bacteremia occurs frequently, there will typically be signs or symptoms suggestive of prosthetic joint infection at presentation. Accordingly, patients in whom a careful history and physical examination do not suggest prosthetic joint infection can be clinically monitored while undergoing treatment for S. aureus bacteremia.
There are several limitations to this study inherent to its retrospective nature. The study relied on data recorded in the medical record. Not every patient had the same evaluation for possible prosthetic joint infection, and misclassification bias is possible, particularly among patients with severe sepsis and early death from S. aureus bacteremia. However, some evaluation for prosthetic joint infection was performed in the majority of the 50 patients without hematogenous prosthetic joint infection, as evidenced by orthopedic surgery consultation (n = 25), plain radiograph (n = 14), or joint aspirate (n = 7). This suggests that the diagnosis of hematogenous prosthetic joint infection was at least considered and may lower the likelihood of misclassification. It is often difficult to differentiate hematogenous prosthetic joint infection from primary prosthetic joint infection initiated at the time of surgery. By using 2 investigator reviews and designating unclear cases as indeterminate, we attempted to minimize this. However, misclassification of the type of prosthetic joint infection is possible. Due to the relatively small sample size, our multivariable analysis was limited. Therefore, it is possible that there were unmeasured confounders underlying some of our observed findings. This may be true with patients with multiple arthroplasties, in whom an inflammatory joint disease might be more likely, and may partially contribute to a higher likelihood of prosthetic joint infection in this group. Finally, our large tertiary, referral center serves many patients with complicated orthopedic histories and multiple prostheses. Accordingly, the absolute frequency of prosthetic joint infection during S. aureus bacteremia may be lower in other settings. However, we suspect that the overall findings and implications of this study are valid.
In summary, these data indicate that hematogenous prosthetic joint infection is frequently seen among patients with S. aureus bacteremia, and that clinical signs or symptoms of prosthetic joint infection are typically present. Prosthetic joint infection is less common in patients with nosocomial S. aureus bacteremia, particularly when only a single arthroplasty is in place.
References
Wisplinghoff H.
Bischoff T.
Tallent S.M.
Seifert H.
Wenzel R.P.
Edmond M.B.
Nosocomial bloodstream infections in US hospitals: analysis of 24,179 cases from a prospective nationwide surveillance study.
Predicting Risk of Endocarditis Using a Clinical Tool (PREDICT): scoring system to guide use of echocardiography in the management of Staphylococcus aureus bacteremia.
Correlation of patient questionnaire responses and physician history in grading clinical outcome following hip and knee arthroplasty. A prospective study of 201 joint arthroplasties.
Funding: This study was supported by the American Heart Association (research grant AHA 12CRP12080058 to MRS) and the Mayo Clinic Center for Clinical and Translational Science (funded by National Institutes of Health Clinical and Translational Science Awards grant UL1 RR024150). The study database was created through REDCap (grant UL1 TR000135).
Conflict of Interest: LMB reports royalty payments (authorship) from UpToDate, Inc (<$20,000) and Editor-in-Chief payments from Massachusetts Medical Society (Journal Watch Infectious Diseases; <$20,000). MRS reports receiving funds from TYRX Inc and Medtronic for prior research unrelated to this study, administered according to a sponsored research agreement (SRA) between Mayo Clinic and study sponsor that prospectively defined the scope of the research effort and corresponding budget; and honoraria/consulting fees from Medtronic, Spectranetics. EFB reports royalty payments (authorship) from UpToDate, Inc (<$5000). Received grant from Merck to conduct a multi-site study on the efficacy of a novel S. aureus vaccine in the prevention of wound infection following complex spine surgery. Remaining authors: No disclosures.
Authorship: All authors had access to the data and played a role in writing the manuscript.
00914-6&id=gr1.jpg){kind=link}
00914-6&id=gr2.jpg){kind=link}