Abstract
Background
Prosthetic joint infection (PJI) is a devastation complication of total joint arthroplasty that can result in poor patient outcomes. Anaerobic organisms make up a small proportion of PJI cases and are much less studied. Studies comparing patient outcomes in anaerobic PJI to outcomes in aerobic PJI are sparse. The purpose of this study was to compare the clinical presentation, duration of antibiotics, type of treatment provided, and final outcome between PJI patients with anaerobic infection and those with aerobic infection.
Methods
This was a retrospective study of 26 patients who underwent treatment for PJI at a tertiary referral center. Eight patients with anaerobic PJI were compared to 18 patients with aerobic PJI in terms of clinical presentation, laboratory values, treatment duration, and functional outcome. Statistical analysis was performed on continuous variables of interest.
Results
The results of our study showed that there are no differences in short term clinical outcomes between PJI patients with cultures positive for anaerobic vs aerobic organisms (38.9% vs 50% successfully treated). Inflammatory markers were higher in the aerobic group and patients in the anaerobic group tended to have fewer medical comorbidities.
Conclusion
PJI caused by anaerobic organisms results in poor patient outcomes similar to infection caused by aerobic organisms. There are some differences in clinical presentation between the two groups that can be explained by the fact that anaerobic organisms are of low virulence and result in indolent infections causing longstanding symptoms.
Keywords: Prosthetic joint infection, Revision, Total knee arthroplasty, Total hip arthroplasty, Anaerobic, Aerobic, Comparative study, Cutibacterium acnes
1. Introduction
One of the most common and devastating complications of total hip (THA) or knee arthroplasty (TKA) is prosthetic joint infection (PJI), occurring at a rate of 1–2%.1, 2, 3 Despite recent advances in sterile techniques, increasing bacterial drug resistance and rising projected demand for joint replacements make PJI a significant problem at present and in the years to come.4 Revision arthroplasty due to PJI poses a significant economic burden. The average cost of revision THA due to PJI was estimated to be around $27,000 per case in a financial analysis study in 2012, and rising healthcare costs have only driven this number higher in the recent years.5
Revision surgery due to PJI also results in significantly adverse patient outcomes. It is associated with longer operating times, increased blood loss, longer length of stay, increased readmissions, and need for further revisions.5, 6, 7 Recent large registry-based studies have also shown that PJI results in an increased mortality risk within 1 year after revision surgery.8,9 Hence, it is crucial to identify factors affecting the complication rate after revision surgery to better predict the prognosis and modify these factors where feasible. While there have been many studies recently comparing outcomes after one-stage versus two-stage revision for PJI, there have been few reports looking at outcomes in anaerobic vs aerobic PJI.10, 11, 12, 13
Anaerobic bacteria make up approximately 3–6% of PJI and include organisms such as Cutibacterium acnes, Clostridium species, Peptostreptococcus species, Actinomyces species, and Bacteroides species.14 PJI due to anaerobic organisms differs from that caused by aerobic organisms in that anaerobic infection does not always result in elevated inflammatory markers.15 Furthermore, cultures take longer to turn positive with anaerobic infections than those caused by aerobic microorganisms.16 Despite these differences in clinical presentation and diagnostic test results, we hypothesized that there would be no differences in overall outcomes in PJI patients infected by anaerobic organisms versus those infected by aerobic organisms. The purpose of this study was to compare the clinical presentation, duration of antibiotics, type of treatment provided, and final outcome between PJI patients with anaerobic infection and those with aerobic infection.
2. Methods
After obtaining institutional review board approval, we conducted a retrospective chart review of patients who received treatment for prosthetic joint infection by a single senior surgeon at our institution between August 1, 2016, and August 1, 2019. Patients who were 18 years or older and had a minimum of 2 months of follow up from last revision surgery were included. Patients with malignancy, incomplete medical records, or pediatric patients were excluded.
A total of 8 patients who met criteria for inclusion had cultures positive for anaerobic organisms. Eighteen patients with PJI who had cultures positive for aerobic organisms and matched with the anaerobic cohort in terms of average age and male-to-female ratio were included for analysis. For each patient included in this study, at least five intraoperative culture specimens were obtained at the time of surgery in a sterile manner and collected in appropriate aerobic and anaerobic specimen collectors. These specimens were taken to the lab for preparation as soon as the collection process was complete to maintain viability of the specimens. Furthermore, all patients included in this study were subjected to aerobic, anaerobic, fungal, and acid fast bacilli (AFB) cultures per protocol at our institution.
All patients who were included for analysis met criteria for PJI based on the 2018 Definition of Periprosthetic Hip and Knee Infection set forth by Parvizi et al.17 Demographic and clinical factors including body mass index (BMI), duration of follow-up, smoking status, history of diabetes mellitus, other medical comorbidities, any complications, type of revision surgery performed, duration of antibiotics, type of antibiotics administered, and functional outcome at final follow-up were recorded for each patient in each cohort. Descriptive and quantitative analysis was performed where appropriate. Student’s t-test was used for continuous variables of interest and sample proportions. Alpha of 0.05 was used as level of significance.
3. Results
3.1. Clinical presentation and demographics
There were 10 patients who underwent TKA and 8 patients who underwent THA in the aerobic cohort. There were 4 patients who underwent TKA and 4 patients who underwent THA in the anaerobic cohort. The average age of patients in the anaerobic cohort was 60.14 (26–85) years and 61.7 (31–81) years in the aerobic cohort. The average BMI was 26.6 (21.1–32.6) Kg/m2 in the anaerobic cohort and 32.6 (25.9–43.2) Kg/m2 in the aerobic cohort. This difference was statistically significant. Even though the aerobic cohort had a greater proportion of patients who were active smokers, diabetic, immunocompromised, had more than two non-musculoskeletal (MSK) co-morbidities, or had prior history of infection than the anaerobic cohort did, these differences were not statistically significant. These results are presented in Table 1.
Table 1.
Clinical presentation and patient characteristics.
| Anaerobic | Aerobic | P-value | |
|---|---|---|---|
| Average Age | 62.3 (26–85) years | 62.0 (31–81) years | 0.9722 |
| Average BMI | 25.6 (18.5–32.6) Kg/m2 | 31.8 (18.5–43.3) Kg/m2 | 0.0132 |
| Female to male ratio | 0.5 | 0.39 | 0.9688 |
| Positive smoking history | 0% | 10.5% | 0.3507 |
| Positive diabetes history | 12.5% | 33.3% | 0.3133 |
| >2 non-MSK medical comorbidities | 37.5% | 68.4% | 0.1473 |
| History of prior infection | 37.5% | 44.4% | 0.7472 |
| Immunocompromised | 0% | 16.7% | 0.2281 |
The most common presenting symptom was pain and swelling of the joint involved followed by wound drainage in both the patient cohorts. For THA, patients in both cohorts also commonly presented with recurrent dislocation prior to being diagnosed with PJI. Three out of the eight patients in the anaerobic cohort had massive osteolysis on radiographs while one patient out of 18 in the aerobic cohort had evidence of loosening on radiographs(see Fig. 1).
Fig. 1.
Differences in patient characteristics between anaerobic and aerobic PJI cohorts.
3.2. Laboratory workup and culture results
The mean white blood cell count was 8.87 (±1.47 SD) x 109/L in the anaerobic cohort and 8.95 (±4.20 SD) x 109/L in the aerobic cohort. Average erythrocyte sedimentation rate (ESR) was mildly elevated in the anaerobic cohort (24.1 ± 26.6 mm/h) and moderately elevated in the aerobic cohort (39.8 ± 29.1 mm/h). Mean C-reactive protein (CRP) was mildly elevated in the anaerobic cohort (17.23 ± 19.5 mg/L) and moderately elevated in the aerobic cohort (84.3 ± 100.0 mg/L). Synovasure was positive in 37.5% of the patients in the anaerobic cohort compared to 44.4% in the aerobic cohort. Average synovial WBC was 3607 (±4476)/uL in the anaerobic cohort and 13,505 (±17,384) in the aerobic cohort. The average neutrophil percent in synovial differential was 57.3 (±28.2) in the anaerobic cohort and 72.5 (±30.9) in the aerobic cohort. All of these differences in the inflammatory marker levels did not reach statistical significance except for the elevation in CRP and synovial WBC. These results and P-values are summarized in Table 2 and Fig. 2.
Table 2.
Differences in inflammatory markers between anaerobic and aerobic cohorts.
| Anaerobic | Aerobic | P-value | |
|---|---|---|---|
| WBC (x 109/L) | 8.87 (6.65–10.61) | 8.95 (1.97–19.5) | 0.9436 |
| ESR (mm/hr) | 24.1 (1–78) | 39.8 (3–110) | 0.1988 |
| CRP (mg/L) | 17.2 (1–47.9) | 84.3 (1–397) | 0.0132 |
| Synovasure positive | 37.5% | 44.4% | 0.7472 |
| Synovial WBC (/uL) | 3607 (78–10,425) | 13,505 (70–60,080) | 0.035 |
| Synovial neutrophil % | 57.3 (13–97) | 72.5 (0–96) | 0.2386 |
Fig. 2.
Inflammatory markers in anaerobic and aerobic cohorts.
The most common organisms isolated on cultures in the aerobic cohort were methicillin-resistant Staphylococcus aureus (MRSA) and Escherichia Coli (E. Coli) followed by methicillin-sensitive Staphylococcus aureus (MSSA), methicillin-resistant Staphylococcus epidermidis (MRSE), and other coagulase-negative Staphylococcus (CNS). The most common organism isolated in the anaerobic cohort was Cutibacterium acnes, followed by Parvimonas. These results are displayed in Fig. 3, Fig. 4.
Fig. 3.
Culture data for aerobic cohort.
Fig. 4.
Culture data for anaerobic cohort.
The average time to diagnosis of PJI from the index arthroplasty procedure was 38.7 (±58.3) months in the aerobic cohort and 96.5 (±141.4) months in the anaerobic cohort. This difference was not statistically significant (P-value = 0.2974).
3.3. Treatment and outcomes
All patients in our study sample underwent either one-stage or two-stage revision surgery for PJI and received long-term oral or intravenous (IV) antibiotics. The average duration of antibiotics was 5.8 (±0.64) weeks for the aerobic group and 6.5 (±3.5) weeks for the anaerobic group. Patients in the anaerobic group were treated with varying combinations of the following antibiotics: amoxicillin, ceftriaxone, nafcillin, penicillin G, and vancomycin. Patients in the aerobic group were treated with varying combinations of the following antibiotics: ampicillin, cefazolin, ceftriaxone, nafcillin, penicillin G, and vancomycin. A total of 37.5% of the anaerobic group and 55.6% of the aerobic group underwent two-stage revision. This difference was not statistically significant (P-value = 0.4035). Furthermore, 100% of the aerobic group and 62.5% of the anaerobic group received IV antibiotics for an extended period (P-value = 0.0068).
The final outcome was classified as “good” if the patient was independently ambulatory and off of all antibiotics at last clinical follow-up. The outcome was classified as “fair” if the patient was ambulating with assistance or still on antibiotic suppression at last clinical follow-up. The outcome was classified as “poor” if the patient expired, required readmission, or experienced a complication such as deep vein thrombosis (DVT) directly as a result of revision surgery for PJI.18 None of the patients in our study sample experienced a poor outcome. There was no statistically significant difference in the number of patients who experienced good outcome in the anaerobic vs aerobic group. These results are summarized in Table 3 and Fig. 5.
Table 3.
Treatment and outcome difference between anaerobic and aerobic cohorts.
| Anaerobic | Aerobic | P-value | |
|---|---|---|---|
| Type of orthopedic intervention for PJI | 2 stage revision – 3 1 stage revision – 5 |
2 stage revision – 10 1 stage revision – 8 |
0.4035 |
| Type of antibiotics received | Oral – 3 IV – 5 |
Oral – 0 IV – 18 |
0.0068 |
| Average duration of follow up (months) | 5.4 (2.5–8) | 13.3 (2.5–27) | 0.0016 |
| Final outcome “good” (%) | 38.9% | 50% | 0.6041 |
Fig. 5.
Treatment and outcome results in both culture cohorts.
4. Discussion
Prosthetic joint infection is a devastating complication of total knee or hip arthroplasty that can result in significant patient morbidity and mortality in addition to posing a substantial economic burden.5, 6, 7, 8, 9,19 The incidence of PJI has been noted to be anywhere from 1 to 2%, with the most common organisms implicated being aerobic bacteria such as Staphylococcus aureus, coagulase-negative Staphylococci, and Enterococci.18,20 While a recent retrospective study of 144 patients by Tsai et al. from Taiwan showed MSSA to be the most common causative organism for PJI, we found that MRSA was the most common causative organism in our aerobic cohort.18 These results reflect how local prevalence of bacteria and hospital antibiograms play an important role in identifying and treating PJI with the appropriate antibiotics.
Anaerobic organisms comprise only 3–6% of PJI.14 Despite their rarity, they are similar to aerobic organisms in terms of their effect on morbidity, complications, and health care costs associated with revision surgery.14 In our study, we found that outcomes after treatment of PJI caused by anaerobic organisms were similar to outcomes after treatment of PJI caused by aerobic organisms (38.9% vs 50% successfully treated). These results are similar to prior research, which showed an overall PJI treatment success rate of 50%.18
Though clinical outcomes between the two cohorts were similar, the clinical presentation and laboratory values differed slightly between the anaerobic and aerobic groups. Patients in the anaerobic cohort had lower average BMI and tended to have fewer medical comorbidities than patients in the aerobic cohort. The aerobic group had a higher average ESR than the anaerobic group at initial presentation (39.8 vs 24.1 mm/h) and higher synovial neutrophil % (72.5 vs 57.3), though these differences did not reach statistical significance. The aerobic group also had a higher average CRP than the anaerobic group (84.3 vs 17.2 mg/L) and higher average synovial WBC (3607 vs 13,505) and these differences were statistically significant. Furthermore, the aerobic group presented with symptoms concerning for PJI an average of 38.7 months after index arthroplasty procedure compared to 96.5 months in the anaerobic group. These results can be explained by the fact that anaerobic organisms like Cutibacterium acnes and Parvimonas are indolent pathogens of low virulence that cause delayed presentation and insidious pain.21
In terms of treatment, 37.5% of the patients in the anaerobic group underwent two-stage revision surgery (explant with antibiotic spacer insertion followed by re-implantation after long-term antibiotics) versus 55.6% of the patients in the aerobic group. This difference can be explained by the fact that PJI was not suspected at the time of surgery in many of the anaerobic cases and diagnosis of PJI was made via intra-operative cultures sent at the time of revision surgery performed for presumed aseptic loosening. Similarly, only 62.5% of the anaerobic cohort received IV antibiotics compared to 100% of the aerobic cohort because the anaerobic cultures turned positive between 4 and 7 days, by which time some of the patients were already discharged from the hospital and chose to take oral antibiotics instead of IV after infection disease consultation. This lack of adequate treatment in the anaerobic group could result in suboptimal outcomes for the patients. Jacobs et al. showed that the incidence of unsuspected PJI was 7.9% in TKA and 12.1% of THA at the time of revision surgery in a retrospective study of 679 patients undergoing surgery for presumed aseptic causes.22 Infection-free survival at two years was inferior in culture-positive TKA patients compared to culture-negative TKA patients while there were no differences in THA patients.22 Furthermore, Portillo et al. showed that median time from primary to revision surgery was much longer in aseptic loosening (7.8 years) than in PJI (2 years) or mechanical failure (1.6 years).23 Cutibacterium acnes was cultured below the established microbiological criteria for positivity in 19% of the revision cases with a working diagnosis of aseptic loosening.23 These published data and our study results suggest that indolent infection by anaerobic organisms should be on the differential for every patient being evaluated for implant loosening, especially if revision surgery is being considered for aseptic causes less than two years from primary arthroplasty.
The indolent nature of the anaerobic organisms could also explain the differences in radiographic findings between the anaerobic and aerobic cohorts. 37.5% of the patients in the anaerobic cohort had massive osteolysis on x-rays prior to revision surgery compared to 5.6% of the patients in the aerobic cohort. Though not statistically significant, there was a trend toward longer average time from index surgery to revision surgery in the anaerobic cohort compared to the aerobic cohort. Therefore, it is the authors’ recommendation to have a low threshold to suspect PJI in patients with massive osteolysis in order to avoid misdiagnosing them with aseptic loosening because, as can be seen from our study results, anaerobic infections can present in a delayed fashion.
One of the limitations of this study was that patient-reported outcome measures (PROMs) such as modified Harris Hip Score or Knee Society Score were not collected or analyzed due to the retrospective design of this study. PROMs are being used with increasing frequency to evaluate patient outcomes after total joint arthroplasty (TJA) and recent research has shown that patient satisfaction may be correlated more strongly with subjective measures than objective measures.22 Thus, future prospective studies evaluating differences in PROMs between patients with anaerobic and aerobic PJI may provide important information for the orthopaedic surgeon treating these patients.
5. Conclusion
Our study showed that outcomes after treatment of PJI caused by anaerobic organisms were similar to outcomes after treatment of PJI caused by aerobic organisms. Clinical presentation and laboratory values differed slightly between the anaerobic and aerobic groups, with a trend toward longer average time to diagnosis of PJI from the index arthroplasty procedure in anaerobic group. Our study results suggest that indolent infection by anaerobic organisms should be on the differential for every patient being evaluated for implant loosening, especially if revision surgery is being considered for aseptic causes less than two years from primary arthroplasty surgery.
CRediT authorship contribution statement
Sravya Vajapey: Funding acquisition, Formal analysis, Data curation, Writing - original draft. Daniel Lynch: Funding acquisition, Formal analysis, Data curation, Writing - original draft. Mengnai Li: Funding acquisition, Formal analysis, Data curation.
Declaration of competing interest
The authors report no conflicts of interest in relation to this work.
References
- 1.Zimmerli W., Trampuz A., Ochsner P.E. Prosthetic-joint infections. N Engl J Med. 2004;351:1645–1654. doi: 10.1056/NEJMra040181. [DOI] [PubMed] [Google Scholar]
- 2.Lentino J.R. Prosthetic joint infections: bane of orthopedists, challenge for infectious disease specialists. Clin Infect Dis. 2003;36:1157–1161. doi: 10.1086/374554. [DOI] [PubMed] [Google Scholar]
- 3.Lidgren L., Knutson K., Stefansdottir A. Infection and arthritis. Infection of prosthetic joints. Best Pract Res Clin Rheumatol. 2003;17:209–218. doi: 10.1016/s1521-6942(03)00002-0. [DOI] [PubMed] [Google Scholar]
- 4.Wimmer M.D., Randau T.M., Friedrich M.J. Outcome predictors in prosthetic joint infections--validation of a risk stratification score for prosthetic joint infections in 120 cases. Acta Orthop Belg. 2016;82:143–148. [PubMed] [Google Scholar]
- 5.Vanhegan I.S., Malik A.K., Jayakumar P., Ul Islam S., Haddad F.S. A financial analysis of revision hip arthroplasty: the economic burden in relation to the national tariff. J Bone Jt Surg Br. 2012;94:619–623. doi: 10.1302/0301-620X.94B5.27073. [DOI] [PubMed] [Google Scholar]
- 6.Schairer W.W., Sing D.C., Vail T.P., Bozic K.J. Causes and frequency of unplanned hospital readmission after total hip arthroplasty. Clin Orthop Relat Res. 2014;472:464–470. doi: 10.1007/s11999-013-3121-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Danish hip arthroplasty register: annual report 2016. https://www.sundhed.dk/content/cms/98/4698_dhr-%C3%A5rsrapport-2016.pdf
- 8.Zmistowski B., Karam J.A., Durinka J.B., Casper D.S., Parvizi J. Periprosthetic joint infection increases the risk of one-year mortality. J Bone Jt Surg Am. 2013;95:2177–2184. doi: 10.2106/JBJS.L.00789. [DOI] [PubMed] [Google Scholar]
- 9.Gundtoft P.H., Pedersen A.B., Varnum C., Overgaard S. Increased mortality after prosthetic joint infection in primary THA. Clin Orthop Relat Res. 2017;475:2623–2631. doi: 10.1007/s11999-017-5289-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Aboltins C., Dowsey M., Peel T., Lim W.K., Choong P. Good quality of life outcomes after treatment of prosthetic joint infection with debridement and prosthesis retention. J Orthop Res. 2016;34:898–902. doi: 10.1002/jor.23089. [DOI] [PubMed] [Google Scholar]
- 11.Wolf M., Clar H., Friesenbichler J. Prosthetic joint infection following total hip replacement: results of one-stage versus two-stage exchange. Int Orthop. 2014;38:1363–1368. doi: 10.1007/s00264-014-2309-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Rieber H., Frontzek A., Jerosch J. Periprosthetic joint infection caused by anaerobes. Retrospective analysis reveals no need for prolonged cultivation time if sensitive supplemented growth media are used. Anaerobe. 2018;50:12–18. doi: 10.1016/j.anaerobe.2018.01.009. [DOI] [PubMed] [Google Scholar]
- 13.Marculescu C.E., Berbari E.F., Cockerill F.R., 3rd, Osmon D.R. Unusual aerobic and anaerobic bacteria associated with prosthetic joint infections. Clin Orthop Relat Res. 2006;451:55–63. doi: 10.1097/01.blo.0000229317.43631.81. [DOI] [PubMed] [Google Scholar]
- 14.Shah N.B., Tande A.J., Patel R., Berbari E.F. Anaerobic prosthetic joint infection. Anaerobe. 2015;36:1–8. doi: 10.1016/j.anaerobe.2015.08.003. [DOI] [PubMed] [Google Scholar]
- 15.Millett P.J., Yen Y.M., Price C.S., Horan M.P., van der Meijden O.A., Elser F. Propionibacterium acnes infection as an occult cause of postoperative shoulder pain: a case series. Clin Orthop Relat Res. 2011;469:2824–2830. doi: 10.1007/s11999-011-1767-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Bhowmick T., Varughese T.A., Arakali S., Boruchoff S.E. Do positive anaerobic culture results affect physicians’ clinical management decisions? Open Forum Infect Dis. 2017;4 doi: 10.1093/ofid/ofw236. ofw236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Parvizi J., Tan T.L., Goswami K. The 2018 definition of periprosthetic hip and knee infection: an evidence-based and validated criteria. J Arthroplasty. 2018;33 doi: 10.1016/j.arth.2018.02.078. 1309-14.e2. [DOI] [PubMed] [Google Scholar]
- 18.Tsai J.C., Sheng W.H., Lo W.Y., Jiang C.C., Chang S.C. Clinical characteristics, microbiology, and outcomes of prosthetic joint infection in Taiwan. J Microbiol Immunol Infect. 2015;48:198–204. doi: 10.1016/j.jmii.2013.08.007. [DOI] [PubMed] [Google Scholar]
- 19.Matthews P.C., Berendt A.R., McNally M.A., Byren I. Diagnosis and management of prosthetic joint infection. BMJ. 2009;338:b1773. doi: 10.1136/bmj.b1773. [DOI] [PubMed] [Google Scholar]
- 20.Jamsen E., Furnes O., Engesaeter L.B. Prevention of deep infection in joint replacement surgery. Acta Orthop. 2010;81:660–666. doi: 10.3109/17453674.2010.537805. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Moran E., Masters S., Berendt A.R., McLardy-Smith P., Byren I., Atkins B.L. Guiding empirical antibiotic therapy in orthopaedics: the microbiology of prosthetic joint infection managed by debridement, irrigation and prosthesis retention. J Infect. 2007;55:1–7. doi: 10.1016/j.jinf.2007.01.007. [DOI] [PubMed] [Google Scholar]
- 22.Jacobs A.M.E., Benard M., Meis J.F., van Hellemondt G., Goosen J.H.M. The unsuspected prosthetic joint infection : incidence and consequences of positive intra-operative cultures in presumed aseptic knee and hip revisions. Bone Joint Lett J. 2017;99-b:1482–1489. doi: 10.1302/0301-620X.99B11.BJJ-2016-0655.R2. [DOI] [PubMed] [Google Scholar]
- 23.Harreld K., Clark R., Downes K., Virani N., Frankle M. Correlation of subjective and objective measures before and after shoulder arthroplasty. Orthopedics. 2013;36:808–814. doi: 10.3928/01477447-20130523-29. [DOI] [PubMed] [Google Scholar]