Infectious Diseases, 2015; Early Online: 1–6
ORIGINAL ARTICLE
Scand J Infect Dis Downloaded from informahealthcare.com by Selcuk Universitesi on 01/31/15 For personal use only.
Successful treatment of Candida osteoarticular infections with limited duration of antifungal therapy and orthopedic surgical intervention
ANDY O. MILLER1,2, MARIA N. GAMALETSOU2,3, MICHAEL W. HENRY1,2, LEEN AL-HAFEZ4, KAISER HUSSAIN1, NIKOLAOS V. SIPSAS2,3, DIMITRIOS P. KONTOYIANNIS5, EMMANUEL ROILIDES2,6, BARRY D. BRAUSE1,2 & THOMAS J. WALSH1,2,7 From the 1Division of Infectious Diseases, Department of Medicine, Weill Cornell Medical Center, New York, 2Center for Osteoarticular Mycoses, Hospital for Special Surgery, New York, NY, USA, 3Medical School, National & Kapodistrian University Athens, Laikon Hospital, Athens, Greece, 4Weill Cornell Medical College in Qatar, Doha, Qatar, 5MD Anderson Cancer Center, Houston, TX, USA, 63rd Department of Pediatrics, Faculty of Medicine, Aristotle University School of Health Science, Hippokration Hospital, Thessaloniki, Greece, and 7Department of Pediatrics and Department of Microbiology and Immunology, Weill Cornell Medical Center, New York, NY, USA
Abstract Background: Current guidelines for treatment of Candida osteoarticular infections (COAIs) recommend a prolonged course of antifungal therapy (AFT) of 6–12 months. Based upon strategies developed at the Hospital for Special Surgery (HSS), we hypothesized that the duration of antifungal therapy may be substantially reduced for management of COAI. Methods: This was a retrospective chart review of cases of COAI treated at the HSS for the past 14 years. COAI was documented by open biopsy and direct culture in all cases. The mean (95% confidence interval, CI) duration of documented follow-up was 39 (16–61) months. Results: Among the 23 cases of COAI, the median age was 62 years (range 22–83 years) with 61% having no underlying condition. Orthopedic appliances, including joint prostheses and fracture hardware, were present in 74% of cases. All patients had COAI as the first proven site of candidiasis. Candida albicans and Candida parapsilosis were the most common species. Hip, knee, foot, and ankle were the most common sites. All patients received aggressive surgical intervention followed by AFT administered for a mean (95% CI) duration of 45 (38–83) days. Systemic AFT consisted principally of fluconazole alone (65%) or in combination with other agents (26%). Adjunctive intraoperative amphotericin B irrigation was used in 35%. Among eight cases of CAOI that required placement of a new prosthetic joint, all were successfully treated. There were no microbiologic failures. Conclusions: Candida osteoarticular infections may be successfully treated with substantially limited durations of AFT when combined with a thorough surgical approach.
Keywords: Candida spp., osteoarticular infections, antifungal therapy, joint prosthesis
Introduction Candida osteoarticular infections (COAIs) are uncommon but debilitating infections. Metastatic bone and joint infection secondary to fungemia in immunocompromised hosts is the best described syndrome; contiguous COAIs, as a complication of orthopedic surgery and trauma, are less commonly encountered. A recent analysis of 207 published
cases of COAIs described a typically insidious onset, a heterogeneous pattern of etiologies, a majority of patients presenting with more than one site of osteomyelitis, and a high frequency of relapsed infections [1]. Recognizing the propensity for relapsed infection, guidelines from the Infectious Disease Society of America (IDSA) recommend that Candida osteomyelitis be treated with 6–12 months of systemic
This study was presented in part, in poster format, at the 53rd ICAAC, Denver, CO, September 2013. Correspondence: Andy O. Miller MD, Hospital for Special Surgery, 535 E 70 Street, New York, NY 10065, USA. Tel: ⫹ 1 212 774 2317. Fax: ⫹ 1 212 774 7412. E-mail: [email protected] (Received 24 July 2014 ; accepted 1 October 2014 ) ISSN 2374-4235 print/ISSN 2374-4243 online © 2014 Informa Healthcare DOI: 10.3109/00365548.2014.974207
Scand J Infect Dis Downloaded from informahealthcare.com by Selcuk Universitesi on 01/31/15 For personal use only.
2
A. O. Miller et al.
antifungal agents in conjunction with surgical debridement [2]. European Society for Clinical Microbiology and Infectious Diseases (ESCMID) guidelines are similar [3]. The Hospital for Special Surgery (HSS) is a leading subspecialty orthopedic hospital serving an international community. The management of COAI at HSS utilizes an aggressive surgical approach by orthopedic specialists with site-specific expertise, which includes aggressive debridement and often amphotericin B deoxycholate (DAmB) irrigations. We hypothesized that this aggressive approach would lead to shorter durations of therapy, reduced relapses, and overall favorable success rates. We further sought to describe the use of local intraoperative irrigation with DAmB in the management of such infections, and to report the outcome in reimplantation of joint prostheses after Candida infection.
Materials and methods Patients This study was approved by the HSS Institutional Review Board in accordance with the ethical standards of the 1964 Declaration of Helsinki and its later amendments. Twenty-three patients were included in this single-center retrospective chart review. Twenty-one original cases of COAI were detected by review of ICD-9 diagnostic codes as follows. Records of all patients between 1988 and 2012 whose list of ICD-9 diagnosis codes included both a fungal diagnosis code and an orthopedic infection diagnosis code were reviewed. Fungal diagnosis codes were 110.0-118, and tissue involvement codes were 711.0-711.99, 730.0-730.99, and 726.0729.9. Of 316 patients with both fungal diagnoses and orthopedic infection diagnoses, 21 patients were identified with CAOI. Subsequently, two more cases lacking correct coding data but known to one of the authors (A.O.M.) were reviewed. Inclusion criteria Cases were included only if direct histopathologic or microbiologic evidence of Candida at the site of orthopedic osteoarticular infection was documented in the inpatient hospital chart or electronic medical record. Definitions Osteoarticular: Related to bone and synovial tissues of joints and bursae. Proven osteoarticular infection: (1) compatible clinical characteristics, (2) consistent radiographic
features, and (3) isolation of Candida in culture and/ or histology from samples of bone tissue, synovial tissue, synovial fluid, or metal hardware obtained by open surgery or percutaneous biopsy [1]. Probable osteoarticular infection: (1) evidence of positive culture of Candida and/or histology from other than bone tissue or metal hardware specimens with compatible clinical and radiologic features. Hematogenous infection: seeding of bone tissue by bloodstream route in the absence of contiguous or direct inoculation. Contiguous infection: presence of an infectious Candida process in close proximity to subsequently infected bone. Direct inoculation: seeding of bone tissue by external trauma, open wound, ulcer, or surgical manipulation. Response to antifungal therapy: complete response, partial response, or failure. Complete response: complete resolution of clinical and radiologic findings of Candida osteomyelitis Partial response: incomplete resolution of clinical and/or radiological findings of osteomyelitis, or partial clinical improvement without availability of follow-up radiological data. Relapse: recurrence of infection after complete or partial response. Failure: death or lack of complete or partial response, despite completion of antifungal therapy. Data collection and analysis Epidemiologic, clinical, demographic, microbiologic, radiographic, and therapy/outcome data were collected, and descriptive statistics were calculated with Microsoft Excel. DAmB irrigation We employed irrigation with 1 L of a 5 mg/dl solution of DAmB, emphasizing the necessity of continuous dependent suction to minimize systemic absorption; no renal injury or other local or systemic toxicity was observed. Subsequent washing with saline or other diluents was not performed. Antibiotic treatment of concurrent bacterial infections In cases of polymicrobial infection, bacterial pathogens were treated with standard antibiotics, typically intravenous cell wall-active agents for gram-positive organisms, and quinolones or beta-lactams for gramnegative organisms, depending on the results of susceptibility testing. Results From 1988 through 2012, 23 identified cases fulfilled predetermined criteria for evaluation, with
Candida osteoarticular infections microbiologic evidence of Candida in deep surgical specimens in all cases. Candida spp. were the only fungi identified.
Table I. Demographic characteristics in 23 cases of Candida osteoarticular infection. Characteristic Median age (years)
Scand J Infect Dis Downloaded from informahealthcare.com by Selcuk Universitesi on 01/31/15 For personal use only.
Demographic characteristics and underlying conditions The median age was 62 years (range 22–83 years; mean 61.8, 95% confidence interval (CI) 55.9–67.7) with a slight male predominance (Table I). Nine patients (39%) had one or more systemic immunocompromising conditions: four (17%) had diabetes mellitus, three (13%) were on chronic immunosuppressive therapy (for solid organ transplant, asthma, or rheumatoid arthritis); active hematologic malignancy, hemodialysis, and human immunodeficiency virus (HIV) infection were present in one patient each. No patients had major risk factors for iatrogenic candidemia, such as prolonged central venous catheterization (CVC), intensive care unit (ICU) care, neutropenia, or total parenteral nutrition. Anatomy and associated sites There were no cases of observed proven or probable candidemia. COAI was the first and only proven site of involvement in 22 patients (Table I; a remaining patient had concurrent funguria). No patient had a prior history of Candida infection. All diagnoses of COAI were made before the initiation of systemic antifungal therapy, which was started before orthopedic surgery in four (17%) cases. Eight patients (35%) presented with periprosthetic joint infections (PJIs), and an additional five (22%) had infections associated with other orthopedic hardware, mostly fracture-associated. Ten (43%) were native infections without orthopedic materials. PJI presented variably as acute purulent infection or chronic infection with loosening of the prosthesis (Figure 1). Consistent with contiguous (rather than bloodborne) infection, no patient had more than one infected orthopedic site. Of five patients with fracture hardware infection, three had femoral fractures. Of eight prosthetic joint infections, Candida was present in prosthetic hips (n ⫽ 5), knees (n ⫽ 2), and shoulder (n ⫽ 1).
3
Male sex Underlying conditions Malignancy, current (lymphoma)* Malignancy, historical (breast, laryngeal) Solid organ transplantation (renal) Prior orthopedic surgery Diabetes* Central vascular catheter Open fracture Neutropenia HIV infection Hemodialysis Pharmacological immunosuppression other than steroids Corticosteroids Total parenteral nutrition Intravenous drug use Intensive care unit Alcohol abuse Prior antifungal agents Orthopedic hardware: Hip prosthesis Knee prosthesis Shoulder prosthesis Fracture hardware Cervical spine fusion No hardware: Olecranon bursa Native knee joint Foot† Wrist joint Native shoulder joint COAI as first proven site of candidiasis Preceding Candida infection
Value
%
61.8 (range 22–83, 95% CI 55.9–67.7) 12
52
1
4
2
9
2
9
22 4 14 1 0 1 1 0
96 17 61 4 0 4 4 0
4 0 0 0 0 2 13 5 2 1 4 1 10 2 2 4 1 1 22
17 0 0 0 0 9 57 22 9 4 17 4 43 9 9 17 4 4 96
0
0
CI, confidence interval; COAI, Candida osteoarticular infection; HIV, human immunodeficiency virus. One patient had two immunocompromising conditions: lymphoma and diabetes mellitus. †One distal tibial post-traumatic osteomyelitis, one calcaneal osteomyelitis after ankle arthrodesis, one transmetatarsal amputation in a hemodialysis patient, and one hallux ostectomy for osteomyelitis.
Clinical microbiology and histopathology CAOIs were diagnosed via preoperative aspiration of fluid (typically under radiologic guidance) or from direct open culture at surgery. Organisms were directly observed on surgical histopathology specimens in only 1 of 23 cases. All patients had monofungal infection; 62% had C. albicans; C. parapsilosis, Candida lusitaniae, and Candida pelliculosa infections
were also encountered (Table II). Also, 55% of cases involved at least one additional bacterial pathogen. Methicillin-resistant Staphylococcus aureus (MRSA) and coagulase-negative staphylococci were present concurrently in 17% each; aerobic gram-negative rods and other bacteria were present in a smaller proportion of cases.
4
A. O. Miller et al.
Scand J Infect Dis Downloaded from informahealthcare.com by Selcuk Universitesi on 01/31/15 For personal use only.
Table III. Clinical manifestations of Candida osteoarticular infection.
Figure 1. Candida prosthetic joint infection with implant loosening. Anteroposterior (AP) and frog lateral of the left total hip replacement shows radiolucency around the femoral component (arrows) and subsidence of the femoral component (arrowheads).
Clinical manifestations All patients presented with local symptoms; systemic symptoms were rare, and only two patients were febrile at presentation (Table III). Sixteen (70%) patients had limitation of function or movement at the site of infection. Serosanguinous or purulent wound drainage from sinus tracts or surgical incisions was noted in nine cases (39%).
All patients were treated with both systemic antifungal agents in conventional dosages and surgical management (Table IV). Irrigation and debridement surgery were performed in all cases, and were supplemented with removal of orthopedic instrumentation Table II. Diagnostic approaches and microbiology of Candida species.
Microbiology Biopsy Percutaneous/closed/guided Intraoperative open culture Culture Candida spp. C. albicans C. parapsilosis C. lusitaniae C. pelliculosa Candida spp. per patient ⫽ 1 Co-cultured organisms MRSA Staphylococcus epidermidis Klebsiella pneumoniae Pseudomonas aeruginosa Enterobacter cloacae ⫹ Chryseobacterium indologenes Mixed oral flora Other fungi
n
%
Local symptoms Edema Erythema Fever Pain Functional limitation Fistula/drainage
23 7 4 3 13 16 9
100 30 17 13 57 70 39
in all of the cases where such material was present. Prosthetic joints were explanted in all cases of periprosthetic joint infection. In three such cases, an antibiotic-laden polymethylmethacrylate (PMMA) spacer was placed in lieu of the infected prosthesis; two spacers had DAmB (at an approximate dose of 15 mg DAmB per gram of PMMA up to a total of 500–750 mg of DAmB) added to the standard vancomycin and aminoglycoside loading. The infected surgical site underwent intraoperative irrigation of a 5 mg/dl DAmB solution, with continuous dependent suction, in 8/23 (35%) of cases. There were no reported complications, local or systemic, in the eight cases of local instillation of DAmB, or in the two cases of PMMA spacers Table IV. Treatment and outcome of Candida osteoarticular infection.
Treatment and outcome
Approach/microbiology
Clinical manifestation
n
%
20 23 23
87 100 100
14 7 1 1 23
61 30 4 4 100
4 4 2 1 1
17 17 9 4 4
1 0
4 0
MRSA, methicillin-resistant Staphylococcus aureus.
Treatment/outcome Intervention Antifungal agents and surgery Only antifungal agents Only surgery Class of systemic antifungal agent(s) used; median duration of treatment (range) Fluconazole monotherapy (range, 21–384 days), mean 45 days Polyene monotherapy (range, 20–77 days), mean 49 days Combination therapy Polyene ⫹ fluconazole (range, 52– 96 days), mean 71 days Echinocandin ⫹ fluconazole (43–70), mean 57 days Echinocandin ⫹ voriconazole; 42 days Local administration of antifungal therapy Polyene irrigation during surgery Use of polyene-loaded PMMA spacer Surgical intervention Surgical irrigation and debridement Removal of hardware/prosthesis Insertion of a PMMA spacer Amputation Total outcome Median duration of therapy (range, 21–294 days) Major functional impairment Relapse of fungal osteoarticular infection PMMA, polymethylmethacrylate.
n
%
23 0 0
100 0 0
15
65
2
9
3
13
2
9
1
4
8 2
35 9
23 17 3 3
100 74 13 13
45 days 6 0
26 0
Scand J Infect Dis Downloaded from informahealthcare.com by Selcuk Universitesi on 01/31/15 For personal use only.
Candida osteoarticular infections containing DAmB. No major toxicities of systemic antifungal therapy were noted. Harmsen et al. reported that concentrations of ⱖ 100 μg/ml of DAmB were cytotoxic to murine osteoblasts and fibroblasts when incubated for 5 h [4]. Consistent with these experimental findings, we saw no clinically overt tissue injury when DAmB 50 μg/ml was used as an intraoperative irrigant. The median total duration of systemic antifungal therapy was 45 days (range 21–294; 95% CI ⫽ 38– 83) and median duration following definitive surgery was 43 days (95% CI ⫽ 38–69). Antifungal regimens included fluconazole monotherapy only (n ⫽ 15, 65%), triazole/echinocandin combination therapy (3, 13%), initial parenteral liposomal AmB (LAmB) transitioned to a triazole (3, 13%), and LAmB alone (2, 9%), respectively. Although 55% of cases had evidence of concomitant bacterial pathogens, no relapse of bacterial infection was observed. Follow-up data were available for 19 patients, in whom the median follow-up time was 17 months (range 1–144 months; mean 38.5, 95% CI ⫽ 16–61). There were major functional impairments in 6 of 19 cases including 3 amputations (1 below the knee, 1 transmetatarsal, and 1 finger), 1 knee arthroplasty, 1 ankle fusion, and 1 humeral head resection. There were no deaths. Discussion This report is distinct in describing the successful treatment of CAOI with limited durations of antifungal therapy (AFT) combined with an aggressive surgical approach by orthopedic specialists with site-specific expertise. Among the 23 cases of CAOI, several common features bear note. These similarities include: relative host immunocompetence in the majority of patients; contiguous inoculation of the pathogen through a surgical or traumatic wound as the likeliest mechanism of infection; prompt orthopedic surgical debridement at the time of diagnosis, performed by site-specific orthopedists at an orthopedic specialty hospital; removal of infected hardware in all relevant cases; finite systemic therapy, often given for a duration significantly shorter than those currently recommended in clinical guidelines; and clinical success, without evidence of recurrent infection. Concomitant bacterial infection was commonplace. We believe that many of these features are common in patients at other centers. C. parapsilosis may be overrepresented in orthopedic device infections because of its propensity for biofilm formation. Although considered less virulent than C. albicans, the concentrations of echinocandins and amphotericin B necessary to eradicate the organism are typically higher. Following adequate removal
5
of hardware, necrotic tissue, and associated biofilm, fluconazole was effective in these patients. While fluconazole has limited activity against biofilm-related C. parapsilosis, the adjunctive surgery likely contributed to removal of biofilm. In only some studies has antifungal duration been systematically characterized [1,5]. By comparison, the present study describes a median duration of 45 days, which we ascribe to aggressive site-specific surgery accompanying systemic antifungal therapy. Further contributing to successful outcome is relative time from symptom onset to surgery; in most cases the interval was acute rather than chronic. Similar to existing reports in the literature, our patients received therapy predominantly with fluconazole and with LAmB [1,5,6]. In the absence of controlled data to the contrary, extended durations of systemic antifungal treatment may be indicated for patients with major immunocompromising conditions and/or fungemia-related orthopedic complications, for whom prognosis is poorer. Of two recent small series of Candida prosthetic joint infections, in which patients were generally treated with two-stage exchange, one reported 100% success rate with a median treatment duration of 6 weeks [7], and a second reported a 50% success rate with a mean of 10 months of therapy [8]. In a third recent series, only one of three patients with twostage exchange and a mean of 5.2 months of therapy had an infection-free outcome [9]. A larger multiinstitutional report of 31 patients with fungal PJI reported a 48% success rate with two-stage prosthesis exchange; duration of antifungal therapy was routinely ⬎ 6 months, although specific data were not reported [6]. Finally, a cohort of 14 patients, 10 of whom underwent one-stage prosthesis exchange, had a mean of 45 days of antifungal therapy, largely fluconazole-based. Nine of the 10 (90%) patients who underwent one-stage exchange were relapse-free at a mean of 7 years’ follow-up [10]. Randomized trials of one- and two-stage prosthesis exchange for PJI (of any type) have not been conducted, but one-stage exchange is a valid approach for fungal PJI. In addition to factors previously discussed, our use of local AFT in selected cases may have played a role. Based upon a premise of delivering high concentrations of DAmB to relatively avascular infected areas, we sought to intensify AFT in cases that would otherwise prove refractory or recurrent. DAmB in high concentrations achieves fungicidal activity against both planktonic and sessile populations within Candida biofilms [11,12]. In candiduria, irrigation of the bladder and of nephrostomy tubes with 5 mg/dl of DAmB is associated with improved fungal clearance [13]. The benefit of intraoperative DAmB irrigation cannot be comparatively evaluated based on
Scand J Infect Dis Downloaded from informahealthcare.com by Selcuk Universitesi on 01/31/15 For personal use only.
6
A. O. Miller et al.
the cases reviewed here, but we believe that its use is, at the very least, a safe and well-tolerated adjunct to therapy for these uncommon infections. PMMA spacers containing DAmB, vancomycin, and an aminoglycoside were used in two cases of PJI, and in another case a spacer without antifungals was employed. Data on the limited elution of DAmB from PMMA may be limited [14]. The role of voriconazole in spacers remains promising albeit incompletely evaluated [15–17]. Other agents such as toremifene may enhance the anti-biofilm potency of current antifungal agents in the future [18]. Infected orthopedic hardware was removed in all of our CAOI cases. This approach is consistent with those of other investigators [6]. Our experience suggests that extended courses of systemic AFT may be unnecessary when adequate specialized surgical debridement and removal of hardware are performed. Limitations of this study include its limited case number, retrospective nature, heterogeneity of the population, site of infection, and surgeon, and the singlecenter experience. In addition, adequate follow-up data were not available for some patients. Only 8 of 23 patients had local DAmB irrigation: many of the patients in this study were treated by a single senior infectious disease consultant who has recommended local DAmB irrigation for many years. Not all orthopedic surgeons agreed to perform this, and local irrigation with DAmB was only recommended if the diagnosis of osteoarticular candidiasis was already known. Despite these limitations we believe that these data advance our understanding of the management of CAOI, particularly in decreasing duration of AFT. In summary, CAOIs present variably, and the duration of systemic AFT may depend on the immune status of the patient and on the adequacy of the orthopedic debridement. We believe that the practices of expert surgical debridement and DAmB irrigation are widely applicable to other institutions. Treatment durations shorter than those recommended in published guidelines appear to be effective and appropriate in the setting of relatively healthy hosts, with nonhematogenous CAOIs, in a specialty orthopedic setting.
Acknowledgments The authors wish to thank Ashlee Reilly, Ruta Petraitiene, and Vidmantas Petraitis for their thoughtful review of the manuscript.
Declaration of interest: No external funding was provided for this study. On behalf of all authors, the corresponding author states that are no commercial relationships or potential conflicts of interest.
References [1] Gamaletsou MN, Kontoyiannis DP, Sipsas NV, Moriyama B, Alexander E, Roilides E, et al. Candida osteomyelitis: analysis of 207 pediatric and adult cases (1970-2011). Clin Infect Dis 2012;55:1338–51. [2] Pappas PG, Kauffman CA, Andes D, Benjamin DK Jr, Calandra TF, Edwards JE Jr, et al. Clinical practice guidelines for the management of candidiasis: 2009 update by the Infectious Diseases Society of America. Clin Infect Dis 2009;48:503–35. [3] Cornely OA, Bassetti M, Calandra T, Garbino J, Kullberg BJ, Lortholary O, et al. ESCMID* guideline for the diagnosis and management of Candida diseases 2012: non-neutropenic adult patients. Clin Microbiol Infect 2012;18(Suppl 7):19–37. [4] Harmsen S, McLaren AC, Pauken C, McLemore R. Amphotericin B is cytotoxic at locally delivered concentrations. Clin Orthop Relat Res 2011;469:3016–21. [5] Slenker AK, Keith SW, Horn DL. Two hundred and eleven cases of Candida osteomyelitis: 17 case reports and a review of the literature. Diagn Microbiol Infect Dis 2012; 73:89–93. [6] Azzam K, Parvizi J, Jungkind D, Hanssen A, Fehring T, Springer B, et al. Microbiological, clinical, and surgical features of fungal prosthetic joint infections: a multi-institutional experience. J Bone Joint Surg Am 2009;91(Suppl 6):142–9. [7] Anagnostakos K, Kelm J, Schmitt E, Jung J. Fungal periprosthetic hip and knee joint infections clinical experience with a 2-stage treatment protocol. J Arthroplasty 2012;27:293–8. [8] Ueng SW, Lee CY, Hu CC, Hsieh PH, Chang Y. What is the success of treatment of hip and knee candidal periprosthetic joint infection? Clin Orthop Relat Res 2013;471: 3002–9. [9] Dutronc H, Dauchy FA, Cazanave C, Rougie C, Lafarie-Castet S, Couprie B, et al. Candida prosthetic infections: case series and literature review. Scand J Infect Dis 2010;42:890–5. [10] Klatte TO, Kendoff D, Kamath AF, Jonen V, Rueger JM, Frommelt L, et al. Single-stage revision for fungal peri-prosthetic joint infection: a single-centre experience. Bone Joint J 2014;96-B:492–6. [11] Prazynska M, Gospodarek E. In vitro effect of amphotericin B on Candida albicans, Candida glabrata and Candida parapsilosis biofilm formation. Mycopathologia 2014;177:19–27. [12] Shuford JA, Rouse MS, Piper KE, Steckelberg JM, Patel R. Evaluation of caspofungin and amphotericin B deoxycholate against Candida albicans biofilms in an experimental intravascular catheter infection model. J Infect Dis 2006;194:710–13. [13] Arthur RR, Drew RH, Perfect JR. Novel modes of antifungal drug administration. Expert Opin Investig Drugs 2004;13: 903–32. [14] Kweon C, McLaren AC, Leon C, McLemore R. Amphotericin B delivery from bone cement increases with porosity but strength decreases. Clin Orthop Relat Res 2011;469:3002–7. [15] Denes E, Fiorenza F, Saint-Marcoux F, Megherbi M, Dupon M, Weinbreck P. Voriconazole stability in cement spacers. Med Mal Infect 2012;42:567–8. [16] Miller RB, McLaren AC, Pauken C, Clarke HD, McLemore R. Voriconazole is delivered from antifungal-loaded bone cement. Clin Orthop Relat Res 2013;471:195–200. [17] Schmidt K, McLaren A, Pauken C, McLemore R. Voriconazole is cytotoxic at locally delivered concentrations: a pilot study. Clin Orthop Relat Res 2013;471:3165–70. [18] Delattin N, De Brucker K, Vandamme K, Meert E, Marchand A, Chaltin P, et al. Repurposing as a means to increase the activity of amphotericin B and caspofungin against Candida albicans biofilms. J Antimicrob Chemother 2014;69:1035–44.
ORIGINAL ARTICLE
Scand J Infect Dis Downloaded from informahealthcare.com by Selcuk Universitesi on 01/31/15 For personal use only.
Successful treatment of Candida osteoarticular infections with limited duration of antifungal therapy and orthopedic surgical intervention
ANDY O. MILLER1,2, MARIA N. GAMALETSOU2,3, MICHAEL W. HENRY1,2, LEEN AL-HAFEZ4, KAISER HUSSAIN1, NIKOLAOS V. SIPSAS2,3, DIMITRIOS P. KONTOYIANNIS5, EMMANUEL ROILIDES2,6, BARRY D. BRAUSE1,2 & THOMAS J. WALSH1,2,7 From the 1Division of Infectious Diseases, Department of Medicine, Weill Cornell Medical Center, New York, 2Center for Osteoarticular Mycoses, Hospital for Special Surgery, New York, NY, USA, 3Medical School, National & Kapodistrian University Athens, Laikon Hospital, Athens, Greece, 4Weill Cornell Medical College in Qatar, Doha, Qatar, 5MD Anderson Cancer Center, Houston, TX, USA, 63rd Department of Pediatrics, Faculty of Medicine, Aristotle University School of Health Science, Hippokration Hospital, Thessaloniki, Greece, and 7Department of Pediatrics and Department of Microbiology and Immunology, Weill Cornell Medical Center, New York, NY, USA
Abstract Background: Current guidelines for treatment of Candida osteoarticular infections (COAIs) recommend a prolonged course of antifungal therapy (AFT) of 6–12 months. Based upon strategies developed at the Hospital for Special Surgery (HSS), we hypothesized that the duration of antifungal therapy may be substantially reduced for management of COAI. Methods: This was a retrospective chart review of cases of COAI treated at the HSS for the past 14 years. COAI was documented by open biopsy and direct culture in all cases. The mean (95% confidence interval, CI) duration of documented follow-up was 39 (16–61) months. Results: Among the 23 cases of COAI, the median age was 62 years (range 22–83 years) with 61% having no underlying condition. Orthopedic appliances, including joint prostheses and fracture hardware, were present in 74% of cases. All patients had COAI as the first proven site of candidiasis. Candida albicans and Candida parapsilosis were the most common species. Hip, knee, foot, and ankle were the most common sites. All patients received aggressive surgical intervention followed by AFT administered for a mean (95% CI) duration of 45 (38–83) days. Systemic AFT consisted principally of fluconazole alone (65%) or in combination with other agents (26%). Adjunctive intraoperative amphotericin B irrigation was used in 35%. Among eight cases of CAOI that required placement of a new prosthetic joint, all were successfully treated. There were no microbiologic failures. Conclusions: Candida osteoarticular infections may be successfully treated with substantially limited durations of AFT when combined with a thorough surgical approach.
Keywords: Candida spp., osteoarticular infections, antifungal therapy, joint prosthesis
Introduction Candida osteoarticular infections (COAIs) are uncommon but debilitating infections. Metastatic bone and joint infection secondary to fungemia in immunocompromised hosts is the best described syndrome; contiguous COAIs, as a complication of orthopedic surgery and trauma, are less commonly encountered. A recent analysis of 207 published
cases of COAIs described a typically insidious onset, a heterogeneous pattern of etiologies, a majority of patients presenting with more than one site of osteomyelitis, and a high frequency of relapsed infections [1]. Recognizing the propensity for relapsed infection, guidelines from the Infectious Disease Society of America (IDSA) recommend that Candida osteomyelitis be treated with 6–12 months of systemic
This study was presented in part, in poster format, at the 53rd ICAAC, Denver, CO, September 2013. Correspondence: Andy O. Miller MD, Hospital for Special Surgery, 535 E 70 Street, New York, NY 10065, USA. Tel: ⫹ 1 212 774 2317. Fax: ⫹ 1 212 774 7412. E-mail: [email protected] (Received 24 July 2014 ; accepted 1 October 2014 ) ISSN 2374-4235 print/ISSN 2374-4243 online © 2014 Informa Healthcare DOI: 10.3109/00365548.2014.974207
Scand J Infect Dis Downloaded from informahealthcare.com by Selcuk Universitesi on 01/31/15 For personal use only.
2
A. O. Miller et al.
antifungal agents in conjunction with surgical debridement [2]. European Society for Clinical Microbiology and Infectious Diseases (ESCMID) guidelines are similar [3]. The Hospital for Special Surgery (HSS) is a leading subspecialty orthopedic hospital serving an international community. The management of COAI at HSS utilizes an aggressive surgical approach by orthopedic specialists with site-specific expertise, which includes aggressive debridement and often amphotericin B deoxycholate (DAmB) irrigations. We hypothesized that this aggressive approach would lead to shorter durations of therapy, reduced relapses, and overall favorable success rates. We further sought to describe the use of local intraoperative irrigation with DAmB in the management of such infections, and to report the outcome in reimplantation of joint prostheses after Candida infection.
Materials and methods Patients This study was approved by the HSS Institutional Review Board in accordance with the ethical standards of the 1964 Declaration of Helsinki and its later amendments. Twenty-three patients were included in this single-center retrospective chart review. Twenty-one original cases of COAI were detected by review of ICD-9 diagnostic codes as follows. Records of all patients between 1988 and 2012 whose list of ICD-9 diagnosis codes included both a fungal diagnosis code and an orthopedic infection diagnosis code were reviewed. Fungal diagnosis codes were 110.0-118, and tissue involvement codes were 711.0-711.99, 730.0-730.99, and 726.0729.9. Of 316 patients with both fungal diagnoses and orthopedic infection diagnoses, 21 patients were identified with CAOI. Subsequently, two more cases lacking correct coding data but known to one of the authors (A.O.M.) were reviewed. Inclusion criteria Cases were included only if direct histopathologic or microbiologic evidence of Candida at the site of orthopedic osteoarticular infection was documented in the inpatient hospital chart or electronic medical record. Definitions Osteoarticular: Related to bone and synovial tissues of joints and bursae. Proven osteoarticular infection: (1) compatible clinical characteristics, (2) consistent radiographic
features, and (3) isolation of Candida in culture and/ or histology from samples of bone tissue, synovial tissue, synovial fluid, or metal hardware obtained by open surgery or percutaneous biopsy [1]. Probable osteoarticular infection: (1) evidence of positive culture of Candida and/or histology from other than bone tissue or metal hardware specimens with compatible clinical and radiologic features. Hematogenous infection: seeding of bone tissue by bloodstream route in the absence of contiguous or direct inoculation. Contiguous infection: presence of an infectious Candida process in close proximity to subsequently infected bone. Direct inoculation: seeding of bone tissue by external trauma, open wound, ulcer, or surgical manipulation. Response to antifungal therapy: complete response, partial response, or failure. Complete response: complete resolution of clinical and radiologic findings of Candida osteomyelitis Partial response: incomplete resolution of clinical and/or radiological findings of osteomyelitis, or partial clinical improvement without availability of follow-up radiological data. Relapse: recurrence of infection after complete or partial response. Failure: death or lack of complete or partial response, despite completion of antifungal therapy. Data collection and analysis Epidemiologic, clinical, demographic, microbiologic, radiographic, and therapy/outcome data were collected, and descriptive statistics were calculated with Microsoft Excel. DAmB irrigation We employed irrigation with 1 L of a 5 mg/dl solution of DAmB, emphasizing the necessity of continuous dependent suction to minimize systemic absorption; no renal injury or other local or systemic toxicity was observed. Subsequent washing with saline or other diluents was not performed. Antibiotic treatment of concurrent bacterial infections In cases of polymicrobial infection, bacterial pathogens were treated with standard antibiotics, typically intravenous cell wall-active agents for gram-positive organisms, and quinolones or beta-lactams for gramnegative organisms, depending on the results of susceptibility testing. Results From 1988 through 2012, 23 identified cases fulfilled predetermined criteria for evaluation, with
Candida osteoarticular infections microbiologic evidence of Candida in deep surgical specimens in all cases. Candida spp. were the only fungi identified.
Table I. Demographic characteristics in 23 cases of Candida osteoarticular infection. Characteristic Median age (years)
Scand J Infect Dis Downloaded from informahealthcare.com by Selcuk Universitesi on 01/31/15 For personal use only.
Demographic characteristics and underlying conditions The median age was 62 years (range 22–83 years; mean 61.8, 95% confidence interval (CI) 55.9–67.7) with a slight male predominance (Table I). Nine patients (39%) had one or more systemic immunocompromising conditions: four (17%) had diabetes mellitus, three (13%) were on chronic immunosuppressive therapy (for solid organ transplant, asthma, or rheumatoid arthritis); active hematologic malignancy, hemodialysis, and human immunodeficiency virus (HIV) infection were present in one patient each. No patients had major risk factors for iatrogenic candidemia, such as prolonged central venous catheterization (CVC), intensive care unit (ICU) care, neutropenia, or total parenteral nutrition. Anatomy and associated sites There were no cases of observed proven or probable candidemia. COAI was the first and only proven site of involvement in 22 patients (Table I; a remaining patient had concurrent funguria). No patient had a prior history of Candida infection. All diagnoses of COAI were made before the initiation of systemic antifungal therapy, which was started before orthopedic surgery in four (17%) cases. Eight patients (35%) presented with periprosthetic joint infections (PJIs), and an additional five (22%) had infections associated with other orthopedic hardware, mostly fracture-associated. Ten (43%) were native infections without orthopedic materials. PJI presented variably as acute purulent infection or chronic infection with loosening of the prosthesis (Figure 1). Consistent with contiguous (rather than bloodborne) infection, no patient had more than one infected orthopedic site. Of five patients with fracture hardware infection, three had femoral fractures. Of eight prosthetic joint infections, Candida was present in prosthetic hips (n ⫽ 5), knees (n ⫽ 2), and shoulder (n ⫽ 1).
3
Male sex Underlying conditions Malignancy, current (lymphoma)* Malignancy, historical (breast, laryngeal) Solid organ transplantation (renal) Prior orthopedic surgery Diabetes* Central vascular catheter Open fracture Neutropenia HIV infection Hemodialysis Pharmacological immunosuppression other than steroids Corticosteroids Total parenteral nutrition Intravenous drug use Intensive care unit Alcohol abuse Prior antifungal agents Orthopedic hardware: Hip prosthesis Knee prosthesis Shoulder prosthesis Fracture hardware Cervical spine fusion No hardware: Olecranon bursa Native knee joint Foot† Wrist joint Native shoulder joint COAI as first proven site of candidiasis Preceding Candida infection
Value
%
61.8 (range 22–83, 95% CI 55.9–67.7) 12
52
1
4
2
9
2
9
22 4 14 1 0 1 1 0
96 17 61 4 0 4 4 0
4 0 0 0 0 2 13 5 2 1 4 1 10 2 2 4 1 1 22
17 0 0 0 0 9 57 22 9 4 17 4 43 9 9 17 4 4 96
0
0
CI, confidence interval; COAI, Candida osteoarticular infection; HIV, human immunodeficiency virus. One patient had two immunocompromising conditions: lymphoma and diabetes mellitus. †One distal tibial post-traumatic osteomyelitis, one calcaneal osteomyelitis after ankle arthrodesis, one transmetatarsal amputation in a hemodialysis patient, and one hallux ostectomy for osteomyelitis.
Clinical microbiology and histopathology CAOIs were diagnosed via preoperative aspiration of fluid (typically under radiologic guidance) or from direct open culture at surgery. Organisms were directly observed on surgical histopathology specimens in only 1 of 23 cases. All patients had monofungal infection; 62% had C. albicans; C. parapsilosis, Candida lusitaniae, and Candida pelliculosa infections
were also encountered (Table II). Also, 55% of cases involved at least one additional bacterial pathogen. Methicillin-resistant Staphylococcus aureus (MRSA) and coagulase-negative staphylococci were present concurrently in 17% each; aerobic gram-negative rods and other bacteria were present in a smaller proportion of cases.
4
A. O. Miller et al.
Scand J Infect Dis Downloaded from informahealthcare.com by Selcuk Universitesi on 01/31/15 For personal use only.
Table III. Clinical manifestations of Candida osteoarticular infection.
Figure 1. Candida prosthetic joint infection with implant loosening. Anteroposterior (AP) and frog lateral of the left total hip replacement shows radiolucency around the femoral component (arrows) and subsidence of the femoral component (arrowheads).
Clinical manifestations All patients presented with local symptoms; systemic symptoms were rare, and only two patients were febrile at presentation (Table III). Sixteen (70%) patients had limitation of function or movement at the site of infection. Serosanguinous or purulent wound drainage from sinus tracts or surgical incisions was noted in nine cases (39%).
All patients were treated with both systemic antifungal agents in conventional dosages and surgical management (Table IV). Irrigation and debridement surgery were performed in all cases, and were supplemented with removal of orthopedic instrumentation Table II. Diagnostic approaches and microbiology of Candida species.
Microbiology Biopsy Percutaneous/closed/guided Intraoperative open culture Culture Candida spp. C. albicans C. parapsilosis C. lusitaniae C. pelliculosa Candida spp. per patient ⫽ 1 Co-cultured organisms MRSA Staphylococcus epidermidis Klebsiella pneumoniae Pseudomonas aeruginosa Enterobacter cloacae ⫹ Chryseobacterium indologenes Mixed oral flora Other fungi
n
%
Local symptoms Edema Erythema Fever Pain Functional limitation Fistula/drainage
23 7 4 3 13 16 9
100 30 17 13 57 70 39
in all of the cases where such material was present. Prosthetic joints were explanted in all cases of periprosthetic joint infection. In three such cases, an antibiotic-laden polymethylmethacrylate (PMMA) spacer was placed in lieu of the infected prosthesis; two spacers had DAmB (at an approximate dose of 15 mg DAmB per gram of PMMA up to a total of 500–750 mg of DAmB) added to the standard vancomycin and aminoglycoside loading. The infected surgical site underwent intraoperative irrigation of a 5 mg/dl DAmB solution, with continuous dependent suction, in 8/23 (35%) of cases. There were no reported complications, local or systemic, in the eight cases of local instillation of DAmB, or in the two cases of PMMA spacers Table IV. Treatment and outcome of Candida osteoarticular infection.
Treatment and outcome
Approach/microbiology
Clinical manifestation
n
%
20 23 23
87 100 100
14 7 1 1 23
61 30 4 4 100
4 4 2 1 1
17 17 9 4 4
1 0
4 0
MRSA, methicillin-resistant Staphylococcus aureus.
Treatment/outcome Intervention Antifungal agents and surgery Only antifungal agents Only surgery Class of systemic antifungal agent(s) used; median duration of treatment (range) Fluconazole monotherapy (range, 21–384 days), mean 45 days Polyene monotherapy (range, 20–77 days), mean 49 days Combination therapy Polyene ⫹ fluconazole (range, 52– 96 days), mean 71 days Echinocandin ⫹ fluconazole (43–70), mean 57 days Echinocandin ⫹ voriconazole; 42 days Local administration of antifungal therapy Polyene irrigation during surgery Use of polyene-loaded PMMA spacer Surgical intervention Surgical irrigation and debridement Removal of hardware/prosthesis Insertion of a PMMA spacer Amputation Total outcome Median duration of therapy (range, 21–294 days) Major functional impairment Relapse of fungal osteoarticular infection PMMA, polymethylmethacrylate.
n
%
23 0 0
100 0 0
15
65
2
9
3
13
2
9
1
4
8 2
35 9
23 17 3 3
100 74 13 13
45 days 6 0
26 0
Scand J Infect Dis Downloaded from informahealthcare.com by Selcuk Universitesi on 01/31/15 For personal use only.
Candida osteoarticular infections containing DAmB. No major toxicities of systemic antifungal therapy were noted. Harmsen et al. reported that concentrations of ⱖ 100 μg/ml of DAmB were cytotoxic to murine osteoblasts and fibroblasts when incubated for 5 h [4]. Consistent with these experimental findings, we saw no clinically overt tissue injury when DAmB 50 μg/ml was used as an intraoperative irrigant. The median total duration of systemic antifungal therapy was 45 days (range 21–294; 95% CI ⫽ 38– 83) and median duration following definitive surgery was 43 days (95% CI ⫽ 38–69). Antifungal regimens included fluconazole monotherapy only (n ⫽ 15, 65%), triazole/echinocandin combination therapy (3, 13%), initial parenteral liposomal AmB (LAmB) transitioned to a triazole (3, 13%), and LAmB alone (2, 9%), respectively. Although 55% of cases had evidence of concomitant bacterial pathogens, no relapse of bacterial infection was observed. Follow-up data were available for 19 patients, in whom the median follow-up time was 17 months (range 1–144 months; mean 38.5, 95% CI ⫽ 16–61). There were major functional impairments in 6 of 19 cases including 3 amputations (1 below the knee, 1 transmetatarsal, and 1 finger), 1 knee arthroplasty, 1 ankle fusion, and 1 humeral head resection. There were no deaths. Discussion This report is distinct in describing the successful treatment of CAOI with limited durations of antifungal therapy (AFT) combined with an aggressive surgical approach by orthopedic specialists with site-specific expertise. Among the 23 cases of CAOI, several common features bear note. These similarities include: relative host immunocompetence in the majority of patients; contiguous inoculation of the pathogen through a surgical or traumatic wound as the likeliest mechanism of infection; prompt orthopedic surgical debridement at the time of diagnosis, performed by site-specific orthopedists at an orthopedic specialty hospital; removal of infected hardware in all relevant cases; finite systemic therapy, often given for a duration significantly shorter than those currently recommended in clinical guidelines; and clinical success, without evidence of recurrent infection. Concomitant bacterial infection was commonplace. We believe that many of these features are common in patients at other centers. C. parapsilosis may be overrepresented in orthopedic device infections because of its propensity for biofilm formation. Although considered less virulent than C. albicans, the concentrations of echinocandins and amphotericin B necessary to eradicate the organism are typically higher. Following adequate removal
5
of hardware, necrotic tissue, and associated biofilm, fluconazole was effective in these patients. While fluconazole has limited activity against biofilm-related C. parapsilosis, the adjunctive surgery likely contributed to removal of biofilm. In only some studies has antifungal duration been systematically characterized [1,5]. By comparison, the present study describes a median duration of 45 days, which we ascribe to aggressive site-specific surgery accompanying systemic antifungal therapy. Further contributing to successful outcome is relative time from symptom onset to surgery; in most cases the interval was acute rather than chronic. Similar to existing reports in the literature, our patients received therapy predominantly with fluconazole and with LAmB [1,5,6]. In the absence of controlled data to the contrary, extended durations of systemic antifungal treatment may be indicated for patients with major immunocompromising conditions and/or fungemia-related orthopedic complications, for whom prognosis is poorer. Of two recent small series of Candida prosthetic joint infections, in which patients were generally treated with two-stage exchange, one reported 100% success rate with a median treatment duration of 6 weeks [7], and a second reported a 50% success rate with a mean of 10 months of therapy [8]. In a third recent series, only one of three patients with twostage exchange and a mean of 5.2 months of therapy had an infection-free outcome [9]. A larger multiinstitutional report of 31 patients with fungal PJI reported a 48% success rate with two-stage prosthesis exchange; duration of antifungal therapy was routinely ⬎ 6 months, although specific data were not reported [6]. Finally, a cohort of 14 patients, 10 of whom underwent one-stage prosthesis exchange, had a mean of 45 days of antifungal therapy, largely fluconazole-based. Nine of the 10 (90%) patients who underwent one-stage exchange were relapse-free at a mean of 7 years’ follow-up [10]. Randomized trials of one- and two-stage prosthesis exchange for PJI (of any type) have not been conducted, but one-stage exchange is a valid approach for fungal PJI. In addition to factors previously discussed, our use of local AFT in selected cases may have played a role. Based upon a premise of delivering high concentrations of DAmB to relatively avascular infected areas, we sought to intensify AFT in cases that would otherwise prove refractory or recurrent. DAmB in high concentrations achieves fungicidal activity against both planktonic and sessile populations within Candida biofilms [11,12]. In candiduria, irrigation of the bladder and of nephrostomy tubes with 5 mg/dl of DAmB is associated with improved fungal clearance [13]. The benefit of intraoperative DAmB irrigation cannot be comparatively evaluated based on
Scand J Infect Dis Downloaded from informahealthcare.com by Selcuk Universitesi on 01/31/15 For personal use only.
6
A. O. Miller et al.
the cases reviewed here, but we believe that its use is, at the very least, a safe and well-tolerated adjunct to therapy for these uncommon infections. PMMA spacers containing DAmB, vancomycin, and an aminoglycoside were used in two cases of PJI, and in another case a spacer without antifungals was employed. Data on the limited elution of DAmB from PMMA may be limited [14]. The role of voriconazole in spacers remains promising albeit incompletely evaluated [15–17]. Other agents such as toremifene may enhance the anti-biofilm potency of current antifungal agents in the future [18]. Infected orthopedic hardware was removed in all of our CAOI cases. This approach is consistent with those of other investigators [6]. Our experience suggests that extended courses of systemic AFT may be unnecessary when adequate specialized surgical debridement and removal of hardware are performed. Limitations of this study include its limited case number, retrospective nature, heterogeneity of the population, site of infection, and surgeon, and the singlecenter experience. In addition, adequate follow-up data were not available for some patients. Only 8 of 23 patients had local DAmB irrigation: many of the patients in this study were treated by a single senior infectious disease consultant who has recommended local DAmB irrigation for many years. Not all orthopedic surgeons agreed to perform this, and local irrigation with DAmB was only recommended if the diagnosis of osteoarticular candidiasis was already known. Despite these limitations we believe that these data advance our understanding of the management of CAOI, particularly in decreasing duration of AFT. In summary, CAOIs present variably, and the duration of systemic AFT may depend on the immune status of the patient and on the adequacy of the orthopedic debridement. We believe that the practices of expert surgical debridement and DAmB irrigation are widely applicable to other institutions. Treatment durations shorter than those recommended in published guidelines appear to be effective and appropriate in the setting of relatively healthy hosts, with nonhematogenous CAOIs, in a specialty orthopedic setting.
Acknowledgments The authors wish to thank Ashlee Reilly, Ruta Petraitiene, and Vidmantas Petraitis for their thoughtful review of the manuscript.
Declaration of interest: No external funding was provided for this study. On behalf of all authors, the corresponding author states that are no commercial relationships or potential conflicts of interest.
References [1] Gamaletsou MN, Kontoyiannis DP, Sipsas NV, Moriyama B, Alexander E, Roilides E, et al. Candida osteomyelitis: analysis of 207 pediatric and adult cases (1970-2011). Clin Infect Dis 2012;55:1338–51. [2] Pappas PG, Kauffman CA, Andes D, Benjamin DK Jr, Calandra TF, Edwards JE Jr, et al. Clinical practice guidelines for the management of candidiasis: 2009 update by the Infectious Diseases Society of America. Clin Infect Dis 2009;48:503–35. [3] Cornely OA, Bassetti M, Calandra T, Garbino J, Kullberg BJ, Lortholary O, et al. ESCMID* guideline for the diagnosis and management of Candida diseases 2012: non-neutropenic adult patients. Clin Microbiol Infect 2012;18(Suppl 7):19–37. [4] Harmsen S, McLaren AC, Pauken C, McLemore R. Amphotericin B is cytotoxic at locally delivered concentrations. Clin Orthop Relat Res 2011;469:3016–21. [5] Slenker AK, Keith SW, Horn DL. Two hundred and eleven cases of Candida osteomyelitis: 17 case reports and a review of the literature. Diagn Microbiol Infect Dis 2012; 73:89–93. [6] Azzam K, Parvizi J, Jungkind D, Hanssen A, Fehring T, Springer B, et al. Microbiological, clinical, and surgical features of fungal prosthetic joint infections: a multi-institutional experience. J Bone Joint Surg Am 2009;91(Suppl 6):142–9. [7] Anagnostakos K, Kelm J, Schmitt E, Jung J. Fungal periprosthetic hip and knee joint infections clinical experience with a 2-stage treatment protocol. J Arthroplasty 2012;27:293–8. [8] Ueng SW, Lee CY, Hu CC, Hsieh PH, Chang Y. What is the success of treatment of hip and knee candidal periprosthetic joint infection? Clin Orthop Relat Res 2013;471: 3002–9. [9] Dutronc H, Dauchy FA, Cazanave C, Rougie C, Lafarie-Castet S, Couprie B, et al. Candida prosthetic infections: case series and literature review. Scand J Infect Dis 2010;42:890–5. [10] Klatte TO, Kendoff D, Kamath AF, Jonen V, Rueger JM, Frommelt L, et al. Single-stage revision for fungal peri-prosthetic joint infection: a single-centre experience. Bone Joint J 2014;96-B:492–6. [11] Prazynska M, Gospodarek E. In vitro effect of amphotericin B on Candida albicans, Candida glabrata and Candida parapsilosis biofilm formation. Mycopathologia 2014;177:19–27. [12] Shuford JA, Rouse MS, Piper KE, Steckelberg JM, Patel R. Evaluation of caspofungin and amphotericin B deoxycholate against Candida albicans biofilms in an experimental intravascular catheter infection model. J Infect Dis 2006;194:710–13. [13] Arthur RR, Drew RH, Perfect JR. Novel modes of antifungal drug administration. Expert Opin Investig Drugs 2004;13: 903–32. [14] Kweon C, McLaren AC, Leon C, McLemore R. Amphotericin B delivery from bone cement increases with porosity but strength decreases. Clin Orthop Relat Res 2011;469:3002–7. [15] Denes E, Fiorenza F, Saint-Marcoux F, Megherbi M, Dupon M, Weinbreck P. Voriconazole stability in cement spacers. Med Mal Infect 2012;42:567–8. [16] Miller RB, McLaren AC, Pauken C, Clarke HD, McLemore R. Voriconazole is delivered from antifungal-loaded bone cement. Clin Orthop Relat Res 2013;471:195–200. [17] Schmidt K, McLaren A, Pauken C, McLemore R. Voriconazole is cytotoxic at locally delivered concentrations: a pilot study. Clin Orthop Relat Res 2013;471:3165–70. [18] Delattin N, De Brucker K, Vandamme K, Meert E, Marchand A, Chaltin P, et al. Repurposing as a means to increase the activity of amphotericin B and caspofungin against Candida albicans biofilms. J Antimicrob Chemother 2014;69:1035–44.
