Abstract
Group A streptococcal (GAS) infections have high morbidity and mortality due to toxin production and tissue invasion. We reviewed all orthopaedic GAS infections at our medical centre between January 2017 and April 2019. Median age was 56 years. At least 60% had a body mass index (BMI) of ≥30 kg/m2. Median Charlson comorbidity score was 3 (range 0–7). Clinical cure at 90 days was achieved in only 62%. All 4 patients with underlying orthopaedic hardware were cured. Toxin-neutralizing antibiotics and intravenous immunoglobulin were underutilized. Our review confirms poor outcomes from GAS orthopaedic infections without optimal management.
Keywords: Group A streptococci (GAS), Streptococcus pyogenes, Necrotizing fasciitis, Osteomyelitis, Hardware infection
1. Introduction
Streptococcus pyogenes or Group A Streptococcus (GAS) are aerobic, non-motile, Gram positive, extracellular bacteria.1 GAS commonly colonizes the nasopharynx and skin. It is a strict human pathogen and contact with respiratory droplets, nasal discharge, and skin lesions are the most frequent modes of transmission.1,2 It is responsible for a wide array of infections, such as impetigo, pericarditis, pharyngitis, pneumonia, and scarlet fever.2,3 The typical incubation period is 1–3 days after exposure.1 Streptococcal toxic shock syndrome, acute rheumatic fever, post-streptococcal glomerulonephritis and necrotizing fasciitis are more severe manifestations of S. pyogenes infection.1,2
Invasive GAS infections of the bone, native joints, prosthetic joints, skin, and soft tissues are often associated with destructive disease requiring aggressive surgical and antimicrobial management.1 Therefore, early recognition of clinical manifestations and microbiological identification is critical. Available data on management of GAS orthopaedic infections is limited, therefore we seek to contribute to existing literature on this important topic. Herein, we describe host risk factors, medical and surgical treatments, and clinical outcomes of GAS orthopaedic infections at our institution and present a comprehensive review of the literature.
2. Material and methods
2.1. Study population and setting
Adult and paediatric patients with culture-positive invasive GAS orthopaedic infections admitted to our academic medical centre in New York City were included in the study. Invasive GAS orthopaedic infection was defined as a native or prosthetic joint infection (PJI), necrotizing fasciitis with associated osteomyelitis, osteomyelitis, septic arthritis, or tenosynovitis. Review of cases was conducted as part of routine Infection Prevention departmental review of institutional group A streptococcus cases; therefore, institutional review board approval was not obtained.
2.2. Study design
We conducted a retrospective review of all cases of invasive GAS orthopaedic infection diagnosed at our institution between January 2017 and April 2019.
2.3. Data collection
A list of all positive GAS cultures was provided by our clinical microbiology laboratory. Cases were reviewed for infection type, site of positive cultures (blood, wound, joint fluid or tissue), demographic data (age, gender, body mass index), medical comorbidities, existing orthopaedic hardware, and immunosuppressive conditions (e.g. diabetes, gammopathy, collagen vascular disorders, and treatment with immunomodulators). We also reviewed surgical procedure type, antimicrobial regimen, route, duration, and use of chronic oral antibiotic suppression. Surgical procedures included: 1) debridement or amputation of native structures, 2) debridement with prosthetic device retention or 3) prosthetic device removal, placement of antibiotic spacer or beads followed by revision surgery.
2.4. Outcomes
The primary outcome was clinical cure at 90 days. Clinical cure was defined as clinical improvement without recurrent or new infection necessitating further surgical or antibiotic treatment within 90 days of index infection. The secondary outcomes were mortality, infectious diseases (ID) consult status, erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) levels before and after treatment (where available) and use of chronic antibiotic suppressionx
3. Results
3.1. Demographics and host risk factors
13 cases (1 paediatric and 12 adult patients) meeting the specified definition of GAS orthopaedic infection were reviewed. The median age of the patients was 56 years (range 3–75). Fifty four percent (n = 7) were female and 62% (n = 8) had a body mass index ≥ 30 kg/m.2 The median Charlson comorbidity index score was 3 (range 0–7) with 54% (n = 8) having a diagnosis of diabetes mellitus. One patient had rheumatoid arthritis (RA) and another had monoclonal gammopathy as a possible predisposing condition (Table 1). Five patients had peripheral vascular disease either alone or with diabetes.
Table 1.
Patient demographics and host factors.
| Patient | Age | Gender | Body Mass Index (kg/m2) | Charlson Comorbidity Index | Medical Comorbiditiesa | |
|---|---|---|---|---|---|---|
| 1 | 73 | F | 35 | 4 | DM, MGUS | |
| 2 | 73 | F | 26 | 7 | DM, PVD | |
| 3 | 53 | M | 32 | 2 | DM | |
| 4 | 56 | M | 30 | 5 | DM, PVD | |
| 5 | 59 | M | 33 | 3 | DM, PVD | |
| 6 | 39 | F | 31 | 2 | DM | |
| 7 | 75 | F | 29 | 4 | none | |
| 8 | 3 | F | 17 | 0 | none | |
| 9 | 37 | F | 26 | 1 | DM | |
| 10 | 56 | M | 27 | 3 | PVD | |
| 11 | 65 | F | 35 | 4 | PVD | |
| 12 | 54 | M | 36 | 3 | RA | |
| 13 | 34 | M | 40 | 0 | none | |
Confirmed by International Classification of Diseases, 10th revision (ICD-10) diagnostic codes in electronic medical record; DM: diabetes mellitus, MGUS: Monoclonal gammopathy of uncertain significance, PVD: peripheral vascular disease, RA: rheumatoid arthritis.
3.2. Diagnosis
Osteomyelitis was the most common musculoskeletal infection type at 62% (n = 8)Necrotizing fasciitis was observed in 31% (n = 4) and all of these patients had associated osteomyelitis confirmed by imaging and/or operative findings. Four patients (31%) had GAS infections associated with previously placed orthopaedic hardware (3 with ankle joint or tibia hardware, 1 with a prosthetic knee joint). All 13 patients had an elevated initial ESR (range 27 to >140 mm/h, reference range <21 mm/h) and CRP levels (range 0.9–42.2 mg/dL, reference range <0.8 mg/dL).
3.3. Surgical and anti-infective treatment
Almost all patients were treated with beta-lactam antibiotics (92%) and at least half received vancomycin at presentation (54%, n = 7). Only 54% (n = 7) received clindamycin or linezolid as toxin-mediating adjunctive therapy. Most patients (77%, n = 10) received at least 2 weeks of antibiotics with a median duration of 3 weeks (range 3 days–6 weeks). Surgical debridement was required in 92% (n = 12), including prosthesis removal in 15% (n = 2). One patient received adjunctive intravenous immunoglobulin for 3 days. Another patient received chronic antibiotic suppression for approximately 9 months while on treatment for rheumatoid arthritis with methotrexate. Infectious diseases consultation was obtained in all but one case, in which the patient left the hospital against medical advice (wound culture from her chronic diabetic foot infection was additionally positive for Pasteurella multocida).
3.4. outcomes
Five patients (38%) were not cured with initial surgical debridement and ultimately required amputation, 1 died of complications of polymicrobial sepsis (Table 2). All patients with infection associated with underlying orthopaedic hardware achieved cure at 90 days regardless of single or multi-stage procedures. One patient with underlying RA received single-stage debridement of a prosthetic knee joint infection followed by 6 weeks of IV ceftriaxone and an additional 9 months of antibiotic suppression with cephalexin and remains infection free at recent follow up. End of treatment levels of ESR, CRP or both were not sent in 62% of cases. Patients requiring amputation or those who expired during index hospitalization had among the highest initial levels of ESR and CRP.
Table 2.
Infection type, treatment strategy, and outcomes.
| Patient | Orthopaedic infection(s) | Date of positive GAS culture | Surgical management | Antibiotic Regimen | Total Antibiotic Duration | ID consult | Initial and final ESR mm/hr) | Initial and final CRP (mg/dL) | Cure at 90 days | Comments |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Right leg necrotizing fasciitis, osteomyelitis; prior tibial ORIF | 7/3/2017 | I&D and fasciotomy | 2 weeks of IV vancomycin/piperacillin-tazobactam/clindamycin or IV penicillin, then 4 weeks of PO amoxicillin/clavulanate | 6 weeks | Y | 110, 39 | 42.4, 3.7 | Yes | IVIG X 3 days |
| 2 | Left hand flexor tenosynovitis | 9/18/2018 | I&D | IV Ceftriaxone and metronidazole for 4 days, then PO amoxicillin/clavulanate for 10 days | 2 weeks | Y | 55, not done | 6.3, not done | Yes | |
| 3 | Right hand flexor tenosynovitis | 10/20/2018 | I&D | IV Ampicillin/sulbactam for 3 days, then PO amoxicillin/clavulanate for 2 weeks | 17 days | Y | 101, 47 | 4.2, 0.8 | Yes | |
| 4 | Left foot necrotizing fasciitis and osteomyelitis | 10/24/2018 | I&D, below knee amputation | IV Vancomycin and piperacillin/tazobactam for 5 days, then IV ceftriaxone andclindamycin for 1 week, then PO linezolid for 17 days | 4 weeks | Y | >140, >140 | 33.4, 9.7 | No | |
| 5 | Left foot osteomyelitis | 10/18/2018 | Foot amputation | PO Linezolid for 4 weeks | 4 weeks | Y | 46, not done | 2.4, not done | No | |
| 6 | Right foot osteomyelitis | 11/30/2018 | None | IV Vancomycin and piperacillin/tazobactam for 3 days | 3 days | N | 72, not done | 1.8, not done | No | Patient left against medical advice; had co-infection with P. multocida |
| 7 | Right foot necrotizing fasciitis and osteomyelitis | 11/24/2018 | Foot amputation | IV Vancomycin and piperacillin/tazobactam for 4 days, then IV ceftriaxone and clindamycin for 3 days, then IV ampicillin/sulbactam for 2 weeks | 3 weeks | Y | 85, not done | 30.8, not done | Yes | Co-infection with MSSA |
| 8 | Left hip and femur septic arthritis and osteomyelitis | 12/30/2018 | I&D | IV Vancomycin and ceftriaxone for 1 week, then PO cephalexin for 6 weeks | 7 weeks | Y | 40, not done | 4.5, <0.5 | Yes | |
| 9 | Right ankle prosthetic joint infection | 12/12/2018 | I&D and prosthesis removal with antibiotic bead placement, followed by revision | IV Ceftriaxone and clindamycin for 1 week, then IV ceftriaxone alone for 5 weeks | 6 weeks | Y | 114, 41 | 17.3, 0.7 | Yes | |
| 10 | Right foot osteomyelitis | 12/3/2018 | Foot amputation | IV Vancomycin and ciprofloxacin for 5 days, then IV cefazolin for 1 week | 12 days | Y | 35, not done | 3.5, not done | No | Co-infection with MSSA |
| 11 | Right leg necrotizing fasciitis and osteomyelitis | 12/2/2018 | Below knee amputation | IV Vancomycin, clindamycin and meropenem for 11 days | 11 days prior to in-hospital death | Y | 27, not done | 31.2, not done | No | Co-infection with P.stuartii, deceased within 90 days from fungemia |
| 12 | Right knee prosthetic joint infection | 2/19/2019 | I&D, one-stage prosthesis revision | IV Ceftriaxone and clindamycin for 1 week, then IV ceftriaxone alone for 5 weeks, then PO cephalexin chronic suppression | 6 weeks, then chronic suppression | Y | 53, 28 | 3.4, 1.5 | Yes | Suppressive cephalexin for ≥1 year due to immuno-suppressive therapy for RA |
| 13 | Right ankle achilles tenosynovitis overlying prior ORIF site | 4/11/2019 | I&D, antibiotic beads placement, followed by skin graft | Vancomycin and piperacillin/tazobactam for 3 days, then cefazolin for 3 days then linezolid for 2 weeks | 3 weeks | Y | 30, not done | 0.9, not done | Yes | Co-infection with MSSA |
IV: intravenous, PO: oral, ESR: erythrocyte sedimentation rate, I&D: incision and drainage, IVIG: intravenous immunoglobulin, MSSA: methicillin-susceptible S. aureus, RA: rheumatoid arthritis.
4. Discussion
GAS clinical manifestations are myriad: pharyngitis, impetigo, cellulitis, scarlet fever, puerperal sepsis, bacteremia, pneumonia, streptococcal toxic shock syndrome, necrotizing fasciitis, endocarditis and less commonly septic arthritis, and osteomyelitis.1 Post-infectious immune-mediated disorders include acute post-streptococcal glomerulonephritis, acute rheumatic fever, and rheumatic heart disease.1 GAS musculoskeletal and orthopaedic infections are less frequently described, therefore we sought to characterize patient demographics, host factors, treatment strategies, and clinical outcomes of patients managed at our institution. Notable features of high-risk patients include age >50 years old, elevated BMI of ≥30 kg/m2, median Charlson comorbidity index score of 3 with more than half of patients with compromised immune status due to diabetes mellitus. Clinical cure at 90 days was achieved in only 62%, however, all 4 patients with GAS infections with underlying orthopaedic hardware were cured.
GAS remains a significant cause of global morbidity and mortality. Incidence of certain invasive isolates in the United Kingdom have increased by more than 10% between 2015 and 2016 according to genomic studies revealing a new emm1 lineage with a 9-fold increased production of streptococcal pyrogenic exotoxins.2 In the United States, the annual incidence is 3.8 cases per 100,000 patients with a case-fatality rate of 11.7% or higher in the presence of septic shock, streptococcal toxic shock syndrome, and necrotizing fasciitis. Per CDC surveillance data, incidence is highest among persons aged ≥65 years or <1 year and among persons of black race. Independent risk factors for mortality include older age, residence in a nursing home, recent surgery, septic shock, necrotizing fasciitis, meningitis, isolated bacteraemia, pneumonia, GAS strains emm type 1 or 3, and underlying chronic illness or immunosuppression.3 2017 national estimates of invasive disease were 7.26 cases per 100,000 persons.4 Our results confirm poor outcomes and lack of clinical cure at 90 days in patients with necrotizing fasciitis and osteomyelitis. These patients also had higher Charlson scores of ≥4 and higher levels of inflammatory markers throughout their course.
Although we did not directly measure virulence factors, it should be mentioned that GAS has several mechanisms for adherence to host cells using cell surface proteins such as lipoteichoic acid, M proteins, pili, fibronectin binding proteins, and others. The nasopharyngeal mucosa and skin are the principal sites of asymptomatic colonization. GAS expresses a multitude of virulence factors to subvert host immune defences. These virulence factors lead to enhanced resistance to phagocytosis, inhibition of complement-mediated functions (deposition, activation, and antibody opsonization), antimicrobial peptide degradation and inactivation, and neutrophil killing. Invasive GAS infections result from its ability to migrate to deep tissues and the bloodstream, trigger the release of proinflammatory mediators and provoke an aggressive immune response from the host. This interplay between host and bacterial factors lead to tissue destruction, vascular leakage and hyperinflammation mediated in part by streptococcal superantigens.1,2
Early and aggressive surgical management along with appropriate antibiotics are critical for optimal outcomes from invasive GAS infections. Failure to recognize impending signs of severe infection leads to delays in treatment, worse outcomes, and higher mortality.5 In a review of necrotizing fasciitis cases due to GAS, features consistently associated with higher mortality were advanced age and debridement after 24 h of admission.5 In our review, the majority of patients required surgical management, often multiple procedures (including amputation) to ensure survival. Of note, all 4 patients in our series with orthopaedic hardware infection were cured regardless of single or multi-stage surgical procedures. All 4 patients had close outpatient ID and orthopaedic follow up at our medical centre to ensure long term cure.
Initial empiric antibiotic treatment of patients with severe infection and sepsis or septic shock should cover GAS as well as Methicillin resistant Staphylococcus aureus, Gram-negative bacilli and anaerobes.6 2014 IDSA practice guidelines recommend intravenous vancomycin or linezolid plus either piperacillin-tazobactam, a carbapenem, or ceftriaxone plus metronidazole.6 Once GAS is identified, antibiotic de-escalation is recommended.6
While susceptibility testing was not performed on our isolates (as per routine laboratory protocol), GAS remains universally susceptible to penicillin, hence this remains the drug of choice for most streptococcal infections. Streptococci are also susceptible to other beta lactams, carbapenems, macrolides, clindamycin, rifampin and vancomycin. The addition of clindamycin to penicillin is considered a more effective regimen than penicillin alone for invasive infections.7 A mouse model of streptococcal myositis indicated higher efficacy of clindamycin compared to penicillin in severe infections possibly explained by the “Eagle effect,” wherein GAS reduces the expression of penicillin binding proteins during the stationary growth phase.8,9
Clindamycin is an important toxin mediator due to its ability to inhibit protein synthesis and suppress bacterial toxin production.9 Additionally, clindamycin activity is not affected by the stage of bacterial growth and it has longer post antibiotic effect than beta-lactams.9 In a retrospective Australian study of 84 patients with invasive GAS infection, use of clindamycin was associated with a lower 30-day mortality (15% vs. 39% in untreated patients).10 Linezolid is a reasonable alternative for clindamycin-resistant GAS isolates, as it has similar toxin neutralizing properties.9 Less than half of our patients (46%) received upfront clindamycin or linezolid as adjunctive therapy, suggesting that this strategy was underutilized, possibly contributing to a low cure rate at 90 days. A paediatric study of 56 invasive GAS infections showed a failure rate of 68% (disease progression despite ≥ 24 h of antibiotics) when cell wall-inhibiting agents (such as beta lactams) were used alone.11 Authors recommend clindamycin plus a beta-lactam antibiotic and surgical management as the most effective strategy.11 In our review, 38% (5/13) of patients who received clindamycin or linezolid at any point in their treatment course achieved cure, while 23% (3/13) did not, all of whom had necrotizing fasciitis with osteomyelitis, or osteomyelitis alone.
The expression of GAS toxins and super antigens can trigger an overwhelming systemic inflammatory response known as streptococcal toxic shock syndrome, which carries a significant mortality risk of 30–70%.9 IVIG is an adjunctive modality for toxic shock syndrome. It provides passive immunity, neutralizes streptococcal toxins and inhibits T cell proliferation and production of inflammatory cytokines.10,12,13 A 2018 meta-analysis showed reduced 30-day mortality (33%–15%) with use of IVIG for streptococcal toxic shock syndrome, with the pooled result across multiple studies being statistically significant.12 Only one of our patients received adjunctive IVIG in addition to clindamycin. She required multiple debridements but did achieve clinical cure at 90 days.
A variety of surgical strategies are employed for management of Streptococcal PJIs but outcomes may vary depending on surgical type and host factors. Streptococcal PJIs account for 4–12% of reported cases and can be managed with debridement, antibiotics, and implant retention (DAIR).14 However, a 2017 multicentre study of 462 streptococcal PJI patients by Lora-Tamayo and colleagues showed higher rates of failure than previously reported. Overall failure was observed in 42% of patients after a median of 62 days of follow up from debridement, and success rates were highly variable among participating sites (44–91%). S. agalactiae was most frequently isolated in 34% and GAS in roughly 8%. Hematogenous infection was noted in 52%, which may have contributed to higher failure rates. Rheumatoid arthritis and immunosuppressive therapy were independent predictors of overall failure, and infection with GAS was significantly associated with early failure within the first 30 days of debridement. Surgical exchange of removable components, early use of adjunctive rifampin, and extended beta-lactam courses of 3 or more weeks (with or without rifampin) were independent predictors of successful outcome.14 Similarly, Fiaux and colleagues' 2016 study evaluated treatment outcomes of 95 episodes of Streptococcal PJI and reported high failures (27.1–32.7%), especially with use of DAIR and in infections involving prosthetic knee joints. Adjunctive use of rifampicin also led to improved outcomes in their study.15 In contrast, Huotari and colleagues state that DAIR remains an ideal surgical option due to increased tolerability, reduced morbidity and recovery time.15 Authors’ 2018 retrospective study of 54 Streptococcal hip and knee PJIs, conducted at a specialized orthopaedic centre demonstrated higher success rates mainly attributed to close surgical and ID follow up and routine exchange of removable parts.16 Despite a high number of infections of hematogenous origin (62%), nearly 98% of original prostheses were retained after a median follow-up of 2.9 years.16
Keller and colleagues evaluated the role of chronic antibiotic suppression in a 2016 study of 89 patients with orthopaedic hardware infections treated with single-stage revision, debridement with retention of hardware, or without surgery, and found that 3-months of suppressive antibiotics may increase the likelihood of treatment success.17 However, outcomes in patients with GAS infections were not specifically reported. In 2019, Shah and colleagues published a multi-centre study of patients with prosthetic knee joint infections treated with DAIR alone vs. DAIR plus chronic oral antibiotic suppression (57 vs. 51 patients, 108 total). Eight percent of infections were caused by Streptococcal species, including GAS (though the exact number of GAS cases was not specified). Patients receiving chronic antibiotic suppression had superior infection free survival without a significant increase in adverse events. Authors note that benefits of chronic suppression beyond 12 months are less clear according to their results. Of note, there were no RA patients in the chronic suppressive group.18 Therefore, less is known about chronic antibiotic suppression for orthopaedic infections in RA patients.
5. Conclusion
In summary, invasive GAS orthopaedic infections necessitate expedited surgical management, targeted antibiotics and adjunctive toxin-neutralizing agents. Despite aggressive surgical and medical management, 5 of 13 (38%) of patients in our series eventually required amputation to control the infection, 4 of which were not considered clinically cured at 90 days (80%). IVIG was likely underutilized, especially in very ill patients, but contributed to a successful outcome in one patient. Long term ID and orthopaedic follow-up may optimize outcomes. Further study is needed to investigate the role of chronic antibiotic suppression in hosts with durable immune suppression.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of competing interest
None.
References
- 1.Walker M.J., Barnett T.C., McArthur J.D. Disease manifestations and pathogenic mechanisms of group a Streptococcus. Clin Microbiol Rev. 2014;27:264. doi: 10.1128/CMR.00101-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Brouwer S., Lacey J.A., You Y., Davies M.R., Walker M.J. Scartlet fever changes its spots. Lancet Infect Dis. 2019 doi: 10.1016/S1473-3099(19)30494-3. Sep 10; [e-pub] [DOI] [PubMed] [Google Scholar]
- 3.Nelson G.E., Pondo T., Toews K.A. Epidemiology of invasive group A streptococcal infections in the United States, 2005-2012. Clin Infect Dis. 2016;63(4):478–486. doi: 10.1093/cid/ciw248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Centres for Disease Control and Prevention Active bacterial core surveillance (ABCs) report: group A Streptococcus. 2000. http://www.cdc.gov/ncidod/dbmd/abcs
- 5.Kalaivani V., Hiremath B.V., Indumathi V.A. Necrotising soft tissue infection-risk factors for mortality. J Clin Diagn Res. 2013;7(8):1662–1665. doi: 10.7860/JCDR/2013/5535.3240. Epub 2013 Aug 01. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Stevens D.L., Bisno A.L., Chambers H.F. Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014 update by the infectious diseases society of America. Clin Infect Dis. 2014;59(2):e10–e52. doi: 10.1093/cid/ciu444. [DOI] [PubMed] [Google Scholar]
- 7.Moore D.L., Allen U.D., Mailman T. Invasive group A streptococcal disease: management and chemoprophylaxis. Paediatr Child Health. 2019;24(2):128–129. doi: 10.1093/pch/pxz039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Stevens D.L., Gibbons A.E., Bergstrom R., Winn V. The Eagle effect revisited: efficacy of clindamycin, erythromycin, and penicillin in the treatment of streptococcal myositis. J Infect. 1988;158(1):23–28. doi: 10.1093/infdis/158.1.23. [DOI] [PubMed] [Google Scholar]
- 9.Stevens D.L., Bryant A.E., Yan S. Invasive group A streptococcal infection: new concepts in antibiotic treatment. Int J Antimicrob Agents. 1994;4(4):297. doi: 10.1016/0924-8579(94)90029-9. [DOI] [PubMed] [Google Scholar]
- 10.Carapetis J.R., Jacoby P., Carville K., Ang S.J., Curtis N., Andrews R. Effectiveness of clindamycin and intravenous immunoglobulin, and risk of disease in contacts in invasive Group A Streptococcal infections. Clin Infect Dis. 2014;59(3):358–365. doi: 10.1093/cid/ciu304. [DOI] [PubMed] [Google Scholar]
- 11.Julie Z., April P., James T. Improved outcome of clindamycin compared with beta-lactam antibiotic treatment for invasive Streptococcus pyogenes infection. Pediatr Infect Dis J. 1999;18(12):1096–1100. doi: 10.1097/00006454-199912000-00014. [DOI] [PubMed] [Google Scholar]
- 12.Parks T., Wilson C., Curtis N., Norrby-Teglund A., Sriskandan S. Polyspecific intravenous immunoglobulin in clindamycin-treated patients with Streptococcal toxic shock syndrome: a systematic review and meta-analysis. Clin Infect Dis. 2018;67(9):1434–1436. doi: 10.1093/cid/ciy401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Waddington C.S., Snelling T.L., Carapetis J.R. Management of invasive group A streptococcal infections. J Infect. 2014;69:S63–S69. doi: 10.1016/j.jinf.2014.08.005. [DOI] [PubMed] [Google Scholar]
- 14.Lora-Tamayo Jaime, Senneville Éric, Alba Ribera. The not-so-good prognosis of streptococcal periprosthetic joint infection managed by implant retention: the results of a large multicentre study. Clin Infect Dis. 15 June 2017;64(12):1742–1752. doi: 10.1093/cid/cix227. [DOI] [PubMed] [Google Scholar]
- 15.Fiaux E., Titecat M., Robineau O. Outcome of patients with streptococcal prosthetic joint infections with special reference to rifampicin combinations. BMC Infect Dis. 2016;16(1):568. doi: 10.1186/s12879-016-1889-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Huotari Kaisa, Vuorinen Markku, Rantasalo Mikko. High cure rate for acute streptococcal prosthetic joint infections treated with debridement, antimicrobials, and implant retention in a specialized tertiary care centre. Clin Infect Dis. 15 October 2018;67(8):1288–1290. doi: 10.1093/cid/ciy335. [DOI] [PubMed] [Google Scholar]
- 17.Keller S.C., Cosgrove S.E., Higgins Y., Piggott D.A., Osgood G., Auwaerter P.G. Role of suppressive oral antibiotics in orthopaedic hardware infections for those not undergoing two-stage replacement surgery. Open Forum Infect Dis. 2016;3(4) doi: 10.1093/ofid/ofw176. ofw176. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Shah NB, Hersh BL, Kreger AM, et al. Benefits and adverse events associated with extended antibiotic use in total knee arthroplasty periprosthetic joint infection [published online April 4, 2019]. Clin Infect Dis. doi:10.1093/cid/ciz261. [DOI] [PMC free article] [PubMed]