header advert
Bone & Joint Open Logo

Receive monthly Table of Contents alerts from Bone & Joint Open

Comprehensive article alerts can be set up and managed through your account settings

View my account settings

Visit Bone & Joint Open at:



Open Access


Three-stage revision arthroplasty for the treatment of fungal periprosthetic joint infection: outcome analysis of a novel treatment algorithm

a prospective study

Download PDF



Fungal periprosthetic joint infections (fPJIs) are rare complications, constituting only 1% of all PJIs. Neither a uniform definition for fPJI has been established, nor a standardized treatment regimen. Compared to bacterial PJI, there is little evidence for fPJI in the literature with divergent results. Hence, we implemented a novel treatment algorithm based on three-stage revision arthroplasty, with local and systemic antifungal therapy to optimize treatment for fPJI.


From 2015 to 2018, a total of 18 patients with fPJI were included in a prospective, single-centre study (DKRS-ID 00020409). The diagnosis of PJI is based on the European Bone and Joint Infection Society definition of periprosthetic joint infections. The baseline parameters (age, sex, and BMI) and additional data (previous surgeries, pathogen spectrum, and Charlson Comorbidity Index) were recorded. A therapy protocol with three-stage revision, including a scheduled spacer exchange, was implemented. Systemic antifungal medication was administered throughout the entire treatment period and continued for six months after reimplantation. A minimum follow-up of 24 months was defined.


Eradication of infection was achieved in 16 out of 18 patients (88.8%), with a mean follow-up of 35 months (25 to 54). Mixed bacterial and fungal infections were present in seven cases (39%). The interval period, defined as the period of time from explantation to reimplantation, was 119 days (55 to 202). In five patients, a salvage procedure was performed (three cementless modular knee arthrodesis, and two Girdlestone procedures).


Therapy for fPJI is complex, with low cure rates according to the literature. No uniform treatment recommendations presently exist for fPJI. Three-stage revision arthroplasty with prolonged systemic antifungal therapy showed promising results.

Cite this article: Bone Jt Open 2021;2(8):671–678.

Take home message

Therapy for fungal periprosthetic joint infection (fPJI) is complex, with low cure rates according to the literature.

No uniform treatment recommendations presently exist for fPJI.

Three-stage revision arthroplasty with prolonged systemic antifungal therapy showed promising results.


Fungal periprosthetic joint infections (fPJIs) are rare but severe complications of total hip arthroplasty (THA) and total knee arthroplasty (TKA). While many publications and therapy regimens for bacterial periprosthetic joint infections (PJI) exist, treatment approaches for periprosthetic fungal infections remain at a very early stage. Diagnosis is often delayed, because fPJI are difficult to detect,1 resulting in a time-lapse before sufficient therapy is initiated. Individual case series, describing various surgical and antifungal therapies, have been reported in the literature.2

It is important to differentiate between therapy approaches, such as debridement, antibiotics and implant retention (DAIR), and one-, two-, or multiple-stage procedures. However, no evidence has been published indicating the superiority of any approach to date. Furthermore, results, as well as the choice of antimycotic agents and the duration of antifungal therapy, differ substantially among various treatment protocols in literature.2

According to the PRO-IMPLANT Foundation (Germany) treatment algorithm, we performed a three-stage revision in combination with long-term systemic antifungal therapy.3 This study aimed to analyze the outcomes of this revision approach as it relates to the paramount objective of septic revision arthroplasty - permanent infection control.


From 2015 to 2018, 623 patients with acute or chronic PJI following TKA or THA were treated in a high-volume infection referral centre. Overall, 20 patients (3.1%) were diagnosed with a chronic fPJI. A total of 18 patients (2.8%) were included in a prospective, single-centre cohort study. In all, eight patients developed fPJI after TKA, and 12 after THA. Two patients with fPJI were treated with long-term suppression due to comorbidity. As a three-stage procedure was not completed, these two patients were not included in the study. Each case intended for a three-stage revision was discussed in a multidisciplinary team (MDT). All 18 patients gave their consent to be treated according to the PRO-IMPLANT Foundation algorithm. Due to the rarity of the diagnosis, no further exclusion criteria were constituted. A minimum follow-up of 24 months was defined. The baseline parameters are shown in Table I.

Table I.

Patient baseline parameters and demographics.

Variable Patient data (n = 18)
Age, yrs, mean (SD); range 72.8 (10.2); 56 to 85
Female sex, n (%) 10 (55.5)
BMI, kg/m2, mean (SD); range 27.9 (6.1); 20 to 38
ASA score, median (SD); IQR 3 (0.78); 2 to 3.5
Previous surgeries, median (SD); IQR 4 (1.98); 3 to 5
CCI, median (SD); IQR 6 (3.1); 2.25 to 8
Left side, n (%) 11 (61.1)
THA, n (%) 11 (61.1)
Smoker, n (%) 3 (16.6)
  1. ASA, American Society of Anaesthesiologists; CCI, Charlson Comorbidity Index; IQR, interquartile range; SD, standard deviation; THA, total hip arthroplasty.

The definition of PJI was based on the European Bone and Joint Infection Society (EBJIS) criteria.4 To provide microbiological evidence of PJI, McNally et al4 defined a single positive sample as “likely infection” for uncommon contaminants (e.g. fungal pathogens). Therefore, a single positive tissue sample was sufficient, as well as microbiological evidence of fungal infection based on analyses of preoperative joint aspiration or sonication fluid for the diagnosis of fPJI.

In the microbiological laboratory, samples were prepared using forceps and scalpel under laminar air flow. Aliquots of the tissue were subsequently placed on different aerobic and anaerobic culture plates, and growth media (blood agar, chocolate agar, Schaedler agar, brain-heart infusion, and Wilkins-Chalgren infusion). Culture was performed under human body temperature conditions (37 °C) for 14 days.

All removed orthopaedic devices were treated in an ultrasonic bath (Bandelin, Germany), sonicate fluid was incubated in blood culture bottles and conventional culturing was performed eventually. Sensitivity and specificity were determined using two-by-two contingency tables.

The primary outcome was defined as successful eradication of infection according to the consensus criteria developed by Diaz-Ledezma et al:4 1) healed wound condition; 2) no infection-related local revision surgery; and 3) absence of PJI-related mortality. Therapy failure was determined as revision owing to septic complications, such as persistent or newly developed PJI. The secondary outcome measure was defined as reimplantation of a prosthesis. The functional outcomes, as measured at the latest follow-up, were included.

Statistical analysis

Data analysis was performed using the Statistical Package for Social Sciences Software (IBM SPSS Statistics version 24; IBM, USA). Descriptive statistical results (mean, median, standard deviation, range, interquartile range and percentage) were recorded to describe comorbidities, complications, and revisions. If a median is stated, the 25th and 75th percentiles are specified in brackets as the interquartile range (IQR). Statistical analysis was performed using Kaplan-Meier estimates and survival curves. The Shapiro-Wilk test was performed to determine normality. Subgroups were compared with the Mann-Whitney U test for non-parametric analysis. Comparative analysis was performed using chi-squared tests for categorical data. The significance level was set at p < 0.05.

Three-stage revision procedure

Treatment of fPJI was performed according to the recommendations of the PRO-IMPLANT Foundation.3 The principles of this algorithm include, but are not limited to, the following key points: no drug holidays prior to reimplantation of the prosthesis; no joint aspiration before reimplantation; biofilm-active therapy only after reimplantation; and antimicrobial therapy for 12 weeks from the date of last positive microbiological evidence (six months for fungal agents). The three-stage exchange procedure must not alter the predetermined interval of six weeks from explantation to reimplantation.

A three-stage revision procedure should be reserved for difficult-to-treat microorganisms (e.g. rifampicin-resistant staphylococci, ciprofloxacin-resistant gram-negative bacteria, and fungi) due to a lack of any effective, biofilm-active treatment option.

In cases of positive preoperative joint aspiration, evidence of fPJI was provided and prosthesis explantation was conducted, combined with extensive debridement and removal of all foreign bodies, and insertion of a custom-made amphotericin B-loaded polymethylmethacrylate (PMMA) spacer.

A scheduled revision was performed three weeks after the index surgery to trigger two key factors for successful treatment of fPJI: additional meticulous debridement to reduce the local burden of fungi; and exchange of the amphotericin B loaded spacer, to ensure continuously high local concentrations of antimycotic agents.5

If fPJI was not detected preoperatively but bacterial PJI was presumed, a vancomycin/gentamicin loaded spacer (COPAL G+V; Heraeus, Germany) was used within the context of a two-stage revision procedure. With detection of microbiological evidence of fPJI in the tissue samples or sonication fluid, revision was scheduled with a switch to an amphotericin B-loaded spacer after three weeks.

As there is no industrial bone cement available, containing antimycotic agents, liposomal amphotericin B was added to commercial bone cement containing vancomycin/gentamycin (COPAL G+V). We added 0.4 g liposomal amphotericin B per 40 g spacer cement, and 0.2 g amphotericin to the fixation cement (Table II). Liposomal amphotericin B results in improved elution compared to regular amphotericin B.6 Postoperatively, a course of intravenous antifungal therapy was initiated with a loading dose of 70 mg caspofungin, which was reduced to 50 mg for patients weighing less than 80 kg from the day after the operation. In cases of an uneventful postoperative period with regular wound healing, antifungal medication was switched to oral administration of 400 mg fluconazole/day. The total duration of antifungal therapy after reimplantation was six months. During that time, outpatients presented once a month and laboratory tests were conducted. In cases of polymicrobial mixed infections, additional biofilm-active antimicrobial therapy was administered.3

Table II.

Local antifungal therapy.

Cement Amphotericin B (liposomal), g Voriconazole, g Gentamicin, g Vancomycin, g
Spacer 0.4 0.4 0.5 to 1 2
Fixation cement 0.2 0.2 0.5 2
  1. Values per 40 g bone cement (polymethylmethacrylate), according to the PRO-IMPLANT Foundation.3


From 2015 to 2018, a total of 18 patients were included in the single-centre study. The median American Society of Anesthesiologists (ASA) physical status classification score was three (IQR 2 to 3.5). The median Charlson Comorbidity Index (CCI) was seven (IQR 2.25 to 8). The diagnosis of fPJI was confirmed by positive preoperative joint aspiration in two patients, and intraoperative tissue samples in 16 patients. Following the EBJIS criteria, a “confirmed infection” was diagnosed in 13, a “likely infection” in five patients.

Polymicrobial infections with proof of bacterial and fungal involvement were detected in seven cases (Table III). The medical history of the cohort showed at least one failed exchange procedure, due to chronic PJI with a median of four operations (IQR 3 to 5) prior to hospital admission.

Table III.

Antimicrobial findings.

Patient no* Preoperative joint aspiration Explantation (n) Scheduled revision (n) Reimplantation (n)§
1 Staphylococcus epidermidis Candida famata (2/6)

S. epidermidis (2/6)
Culture negative (0/5) Culture negative (0/5)
2 Candida parapsilosis C. parapsilosis (2/9) Culture negative (0/5) Culture negative (0/3)
3 Candida albicans (2/4) C. albicans (1/3) Culture negative (0/5)
4 Candida famata (3/9)

Citrobacter koseri (9/9)
C. famata (2/5) Culture negative (0/5)
5 Candida tropicalis (1/4)

S. epidermidis (2/4)
Culture negative (0/4) Culture negative (0/3)
6 Candida parapsilosis (1/4)

Staphylococcus caprae caprae (3/4)
Culture negative (0/4) Culture negative (0/5)
7 C. albicans (1/4) C. albicans (3/3) Culture negative (0/5)
8 C. parapsilosis (2/6) Culture negative (0/3) Culture negative (0/7)
9 Candida glabrata (2/7)

Escherichia coli (4/7)
Culture negative (0/7) Culture negative (0/7)
10 C. famata C. famata (1/4) Culture negative (0/4) Culture negative (0/3)
11 C. parapsilosis (1/3) Culture negative (0/5) Culture negative (0/3)
12 C. albicans (2/5) Culture negative (0/4) Culture negative (0/3)
13 C. albicans (3/5) C. albicans (3/4) Culture negative (0/4)
14 S. epidermidis C. albicans (1/5)

S. epidermidis (3/5)
C. albicans (3/4) Culture negative (0/5)
15 C. parapsilosis (2/5)

Streptococcus sanguinis (5/5)
Culture negative (0/3) Culture negative (0/7)
16 C. albicans (1/7) Culture negative (0/5) Culture negative (0/5)
17 Alternaria infectoria (2/5) Culture negative (0/5) Culture negative (0/6)
18 C. albicans (1/4) Culture negative (0/3) Culture negative (0/4)
  1. *

    Patients 2 and 10 received an amphotericin B spacer during the first operation.

  1. Only positive results listed.

  1. First number within parentheses indicates positive test results; second number indicates the absolute number of samples collected.

  1. §

    Patients 7, 9, and 13 received a cementless modular arthrodesis; patients 12 and 15 underwent a Girdlestone procedure.

Four patients presented with a fistula and three patients with a soft tissue defect, requiring additional reconstructive surgery. All patients showed microbiological evidence of PJI, and suffered from specific complaints due to chronic PJI with a relevant loss of quality of life.

The mean follow-up was 35 months (25 to 54). In all, 16 patients (88.8%) maintained revision-free survival, as defined by the consensus criteria by Diaz-Ledezma et al.7 The follow-up details are presented in Table IV and the Kaplan-Meier estimator in Figure 1. At the latest follow-up, one patient required an additional two-stage revision owing to a bacterial PJI. One patient deceased due to septic multi-organ failure despite carrying out a Girdlestone resection arthroplasty as an emergency procedure. No patient was lost to follow-up.

Table IV.

Therapy protocol.

Patient no. Entity Age, yrs CCI Previous surgeries* Clinical feature Reconstructive surgery Follow-up, mnths Complication Primary outcome Secondary outcome
1 THA 85 10 2 44 Bacterial PJI after 24 months Two-stage exchange rTHA
2 TKA 66 2 1 35 None IFS rTKA
3 THA 81 9 3 42 AKI IFS rTHA
4 THA 67 2 3 27 None IFS rTHA
5 TKA 81 8 5 28 CKD dialysis IFS rTKA
6 THA 81 7 1 Fistula 41 AKI IFS rTHA
7 TKA 56 1 5 Soft-tissue defect Gastrocnemius flap 54 None IFS Cementless KA
8 THA 78 7 1 Fistula 35 Dislocation

(closed reduction)
9 TKA 66 5 6 Soft-tissue defect Anterolateral thigh flap 51 None IFS Cementless KA
10 THA 81 10 5 26 Nausea IFS rTHA
11 TKA 57 2 4 25 Nnausea IFS rTKA
12 THA 82 7 7 Fistula 26 Persistent PJI Patient deceased Girdlestone
13 TKA 79 9 4 Soft-tissue defect Gastrocnemius flap 30 None IFS Cementless KA
14 THA 78 3 3 26 Nausea IFS rTHA
15 THA 60 8 7 Fistula 39 None IFS Girdlestone
16 TKA 71 8 4 29 None IFS rTKA
17 THA 83 7 3 37 Nausea IFS rTHA
18 THA 57 3 6 35 None IFS rTHA
  1. *

    Primary arthroplasty was defined as index surgery (0).

  1. Successful treatment after two stage-exchange.

  1. AKI, acute kidney injury; CCI, Charleson Comorbidity Index; CKD, chronic kidney disease; IFS, infection-free survival; KA, knee arthrodesis; PJI, periprosthetic joint infection; rTHA, revision total hip arthroplasty; rTKA, revision total knee arthroplasy; THA, total hip arthroplasty; TKA, total knee arthroplasty.

Fig. 1 
          Kaplan-Meier survival estimates.

Fig. 1

Kaplan-Meier survival estimates.

In 13 patients (72.2%), revision joint arthroplasty was performed after a mean of 119 days (55 to 202). In two patients, permanent Girdlestone resection arthroplasty was performed, and three patients received a cementless, modular knee arthrodesis as a salvage procedure. A detailed MDT audit was preceded for the selection of any salvage procedure.

Overall, 17 patients received a regular postoperative course of oral fluconazole from day ten and for another six months after reimplantation. Antifungal therapy was initiated with the intravenous administration of caspofungin. Fluconazole-resistant Candida albicans was detected in one patient. As no appropriate oral antimycotic therapy was available, outpatient parenteral antibiotic therapy (OPAT) with caspofungin was initiated for six months.8 Seven patients with confirmed polymicrobial bacterial and fPJIs received additional antibacterial therapy, starting from the day of explantation. A biofilm-active antimicrobial agent was administered for six weeks after reimplantation.3

Adverse reactions, due to long-term antimicrobial and antifungal therapy, were monitored on a regular basis. Two patients showed signs of acute kidney injury. One patient developed chronic kidney disease, with the need for intermittent haemodialysis. During outpatient presentation, four patients complained about nausea. The blood tests during routine monitoring showed no remarkable pathological findings. One patient was readmitted after early postoperative THA dislocation. Conservative treatment was initiated after closed reduction.

At the last follow-up examination, nine out of 13 patients with revision joint arthroplasty were able to walk independently. Three patients used forearm crutches and reported a mean daily walking distance of about 500 metres. One patient reported sufficient mobility for short distances at home. One patient with a permanent Girdlestone situation was able to transfer from the bed to a wheelchair independently, but could not walk short distances without major limitations. The other patient subjected to a Girdlestone procedure was immobile and subsequently died due to multiorgan failure. All three patients with knee arthrodesis depended on the use of forearm crutches, and were able to walk short distances without technical aids.


Implementing standardized treatment algorithms for PJI has led to a significant improvement of the outcome in this integral aspect of orthopaedic surgery.3,9 The principles of treatment for bacterial PJI cannot be equated with those for fPJI owing to the differences in difficult-to-treat fungal organisms, lower levels of traceability, and a lack of relevant expertise.1 Treatment algorithms for fPJI are a hot topic with only marginal evidence available at present. Different surgical and antifungal therapy regimens show divergent results, which emphasizes the need for further investigations (Table V).

Table V.

Literature on fungal periprosthetic joint infections.

Reference Patients, n Period Surgical procedure Follow-up, mnths Infection eradication, %
Garcia-Oltra et al 201110 3 2002 to 2010 DAIR 31 0
Two-stage revision 31 28
Anagnostakos et al 201211 7 2004 to 2009 Two-stage revision

Fluconazole, six weeks postoperatively
28 100
Ueng et al 201312 7 2000 to 2010 Two-stage revision 31 28
Geng et al 201613 8 2000 to 2012 Two-stage revision

Fluconazole, six weeks postoperatively
53 77.5
Ji et al 20179 11 2004 to 2014 One-stage revision 60 63
Kuo et al 201814 29 1999 to 2014 DAIR 60 28.6
One-stage revision 60 33.3
Two-stage revision 60 46.3
Gao et al 201815 17 2000 to 2015 Two-stage revision 65 72.2
Escola-Verge et al 201816 20 2003 to 2015 DAIR 27
15 Implant revision 67
Kim et al 201817 9 2001 to 2016 Two-stage revision 100
Brown et al 201818 31 1996 to 2014 DAIR, one- and two-stage revision 24 55
Theil et al 201919 26 2009 to 2017 Two-stage revision 33 38.5
Current Study 18 2015 to 2019 Three-stage revision

Fluconazole, six months postoperatively
25 88.8
  1. DAIR, debridement, antibiotics and implant retention.

The scope of our study was to contribute scientific data to the controversial discussion on definition and modern treatment algorithms in the field of fPJI. Literature does not show any superiority of either approach (one-, two-, or three-stage). Hence, our presented data suggest the effectiveness of this procedure, in particular after multiple revisions before. A three-stage procedure is burdensome, but the value of a best possible outcome should be considered higher than the potential stress of one additional scheduled revision compared to two-stage revisions. Due to the risk of false negative samples, especially in the treatment of difficult-to-treat organisms (such as fungi), and the lack of a reliable intraoperative assessment for those, it is a paramount objective of the three-stage approach to minimize the risk of persistent fungal colonisation that lead to therapy failure. This safety for all patients must be counterbalanced with potentially one odd operation for only some patients.

As this was a short-term study, the follow-up seems comparable to other studies.18 The recruitment interval from 2015 to 2018 was shorter than that of other studies. Therefore, regular outpatient presentation should be established to ensure continuous re-evaluation.

Diagnosis of fPJI can be exceedingly difficult, particularly during the early stages of disease, and requires a thorough work-up. Patients often complain about only few symptoms, similar to those of low-grade PJI.16 Patients with several comorbidities are particularly susceptible to fPJI.13 Various risk factors have been identified, such as immunosuppression, obesity, diabetes, number of previous revision surgeries, as well as long-term antibiotic treatment.2,13 Analyses of the baseline parameters of the patients in the present study support these findings.

The literature shows a high rate of mixed bacterial and fungal PJIs, which impede the effects of appropriate diagnostics and consistent therapy. These findings are consistent with the results of the present study. Therefore, preoperative joint aspiration should be mandatory during the first diagnostic examination if bacterial of fungal PJI is presumed. Mixed infections are associated with markedly worse outcomes in the literature. Sidhu et al20 reported a revision-free survival rate of 38% among patients with mixed bacterial infections. In contrast, only one treatment failure occurred in the mixed infections cohort of this study. It is assumed that the additional antimicrobial, biofilm-active therapy was highly effective,3 but we are aware that these are only short-term follow-up results.

The spectrum of fungal pathogens detected in the present study was comparable with that of previous publications, particularly the frequency of infections with Candida albicans and Candida parapsilosis.21 Infections with mould fungal species, as we observed in one patient with Alternaria infectoria, are rare and have only been reported in two previous case studies.22

No uniform treatment recommendations or diagnostic standards exist for fPJI. The International Consensus Meeting on Musculoskeletal Infection 2018 (Philadelphia, USA) discussed fPJI in four questions, with low evidence and vague therapy recommendations.23 The recently published EBJIS definition by McNally et al4 facilitates the use of defined criteria in a consistent algorithm.4 Accordingly, a single positive microbiological sample of an uncommon or highly virulent organism should be considered as a likely infection. Contamination with these pathogens is unlikely and the consequences of a missed infection might be devastating. In accordance, the PRO-IMPLANT Foundation categorizes fungal agents as highly virulent, though with a comparably low level of evidence.3

The overall need for a future precise, well-accepted definition of fPJI must be emphasized. An important step in this direction was made by publication of the EBJIS criteria for PJI definition.4 We acknowledged that the lack of evidence concerning generally adopted definition criteria for fPJI is a limitation of the present study.

Currently, neither refined diagnostic techniques, nor any other tools, are available to improve the determination of fPJI. In future investigations, process sequencing of the internal transcribed spacer segment, for fungal pathogens specifically, could enhance the diagnostic capability.24

Unlike for bacterial PJI, there are no histological standards defined for either classification of fPJI. The histopathological criteria for bacterial PJI, established by Morawietz et al25 are not explicitly defined for fPJI. It remains unclear whether histological observations about bacterial infections can be applied to fungal infections.23 For this reason, histology was no dedicated focus of our study protocol, although this feature might be of interest for further investigations.

Antimicrobial therapy as a cornerstone of successful PJI treatment is indispensable in every treatment algorithm.3 However, recommendations regarding the appropriate duration differ and can range from six to 52 weeks.26 Meta-analyses have identified improved eradication of infection with prolonged systemic therapy from three to six months.12 This is consistent with our findings, which showed high rates of eradication following six months of antifungal treatment after three-stage revision arthroplasty.

The second key factor of antifungal treatment, besides systemic therapy, is adequate local therapy. When adding antifungal agents to the bone cement mixture, certain characteristics must be considered. The mixing ratio for bone cement differs among spacers and the cement used during reimplantation. High drug doses affect the mechanical stability of PMMA, particularly if two or more drugs are used. At the time of revision arthroplasty, antibiotics should not comprise more than 10% to 15% of the cement mass.3 The lower dose of added antimycotics and the third-generation cementing technique for the fixation cement result in superior mechanical properties.5

Considering the microbiological findings, as well as the point in time at which the fPJI diagnosis was made, it became clear that if a fungal infection was diagnosed preoperatively and an antimycotic PMMA spacer was used initially, no persistent infection (0%) was detected in subsequent operations. In the event that fPJI was diagnosed in tissue samples acquired during explantation without preoperative diagnosis of fungal infection, and implantation of a spacer without antifungal agents during the index operation, fPJI was detected in five of 16 (31.2%) spacer exchanges (Table III). However, due to the small study group that difference was not significant (p = 0.51, Fisher's exact test).

The overall importance of systemic and local antifungal therapy is underlined in the present study, which is consistent with the findings of previous studies.1 Furthermore, the present findings reveal the essential advantage of a three-stage revision strategy. Repeated surgical debridement, combined with a scheduled spacer exchange, could facilitate continuous delivery of the highest local drug concentrations with optimized release kinetics and without systemic side-effects.5 Furthermore, certain disadvantages may be associated with a three-stage procedure, such as increased perioperative and postoperative risks, including mortality and adverse drug effects, due to prolonged antifungal and antibacterial therapy. Indications for additional scheduled surgery must be justified by evidence of improved outcomes compared with those of one- or two-stage procedures.14,18 Depending on the antimicrobial therapy prescribed, advanced outpatient care is required to monitor potential complications.

Indications for any salvage procedure were closely evaluated by a MDT and confirmed after a detailed explanatory meeting and risk assessment regarding the outcome.27 Salvage therapy should only be considered in cases of high perioperative risks or patient refusal of an eradication strategy. First-line salvage therapy in patients with chronic fPJI, in our opinion, must be long-term antifungal suppression with fluconazole or voriconazole.3 The establishment of a persistent fistula should be the last option.27

Knee arthrodesis was only performed in case of complete insufficiency of the extensor apparatus, in combination with an extensive compromise of the soft tissue envelope and additional reconstructive surgery for soft tissue coverage.

A permanent Girdlestone situation only was feasible if immobility must be tolerated, had previously existed, or the patient denied consent to further surgical treatment. The latter is the case in the investigated cohort. Both Girdlestone patients denied further multiple-stage revision for personal reasons.

There are limitations to the current study. Foremost, cohorts investigating fungal PJI are limited by patient numbers. We identified 18 patients, thus the strength of the data may be weakened. However, this study is one of the largest cohorts reported with a standardized treatment algorithm for fungal PJI.

Following the EBJIS criteria, we considered one positive sample to be sufficient for a likely infection.4 The lack of generally accepted guidelines concerning diagnosis and treatment of fPJI must be emphasized. The focus was to evaluate the rate of successful infection eradication. Patient-reported outcome measures need to be addressed in further investigations.

In conclusion, treatment of fPJI is a challenge that requires a multidisciplinary standardized team approach with individual case analyses for optimized outcomes. Three-stage revision arthroplasty, combined with local and systemic antifungal therapy, shows promising results and is associated with high revision-free survival rates in a short-term follow-up. Long-term results are necessary to confirm the present findings.

Correspondence should be sent to Hinnerk Baecker. E-mail:


1. Nace J , Siddiqi A , Talmo CT , Chen AF . Diagnosis and management of fungal periprosthetic joint infections . J Am Acad Orthop Surg . 2019 ; 27 ( 18 ): e18e804 : e804 e818 . Crossref PubMed Google Scholar

2. Cobo F , Rodriguez-Granger J , Sampedro A , Aliaga-Martinez L , Navarro-Mari JM . Candida prosthetic joint infection: a review of rreatment methods . J Bone Jt Infect . 2017 ; 2 ( 2 ): 114 121 . Google Scholar

3. Izakovicova P , Borens O , Trampuz A . Periprosthetic joint infection: current concepts and outlook . EFORT Open Rev . 2019 ; 4 ( 7 ): 482 494 . Crossref PubMed Google Scholar

4. McNally M , Sousa R , Wouthuyzen-Bakker M , Chen AF , Soriano A , Vogely HC , et al. The EBJIS definition of periprosthetic joint infection . Bone Joint J . 2021 ; 103-B ( 1 ): 18 25 . Crossref PubMed Google Scholar

5. 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 ( 11 ): 3002 3007 . Crossref PubMed Google Scholar

6. Houdek MT , Greenwood-Quaintance KE , Morrey ME , Patel R , Hanssen AD . Elution of high dose amphotericin B deoxycholate from polymethylmethacrylate . J Arthroplasty . 2015 ; 30 ( 12 ): 2308 2310 . Crossref PubMed Google Scholar

7. Diaz-Ledezma C , Higuera CA , Parvizi J . Success after treatment of periprosthetic joint infection: a Delphi-based international multidisciplinary consensus . Clin Orthop Relat Res . 2013 ; 471 ( 7 ): 2374 2382 . Crossref PubMed Google Scholar

8. Baecker H , Gessmann J , Hanusrichter Y , Zahn PK , Schildhauer TA , Frieler S . Outpatient parenteral antibiotic therapy (OPAT) with peripherally inserted central catheter in patients with periprosthetic joint infection . Z Orthop Unfall . 2019 ; 157 ( 5 ): 510 514 . Crossref PubMed Google Scholar

9. Ji B , Zhang X , Xu B , Guo W , Mu W , Cao L . Single-stage revision for chronic fungal periprosthetic joint infection: An average of 5 years of follow-up . J Arthroplasty . 2017 ; 32 ( 8 ): 2523 2530 . Crossref PubMed Google Scholar

10. Garcia-Oltra E , Garcia-Ramiro S , Martinez JC , Tibau R , Bori G , Bosch J . Prosthetic joint infection by Candida spp . Rev Esp Quimioter . 2011 ; 24 ( 1 ): 37 41 . Google Scholar

11. 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 ( 2 ): 293 298 . Crossref PubMed Google Scholar

12. Ueng SWN , Lee C-Y , Hu C-chien , Hsieh P-H , Chang Y . What is the success of treatment of hip and knee candidal periprosthetic joint infection? Clin Orthop Relat Res . 2013 ; 471 ( 9 ): 3002 3009 . Crossref PubMed Google Scholar

13. Geng L , Xu M , Yu L , Li J , Zhou Y , Wang Y , et al. Risk factors and the clinical and surgical features of fungal prosthetic joint infections: A retrospective analysis of eight cases . Exp Ther Med . 2016 ; 12 ( 2 ): 991 999 . Crossref PubMed Google Scholar

14. Kuo FC , Goswami K , Shohat N , Blevins K , Rondon AJ , Parvizi J . Two-Stage Exchange Arthroplasty Is a Favorable Treatment Option Upon Diagnosis of a Fungal Periprosthetic Joint Infection . J Arthroplasty . 2018 ; 33 ( 11 ): 3555 3560 . Google Scholar

15. Gao Z , Li X , Du Y , Peng Y , Wu W , Zhou Y . Success Rate of Fungal Peri-Prosthetic Joint Infection Treated by 2-Stage Revision and Potential Risk Factors of Treatment Failure: A Retrospective Study . Med Sci Monit . 2018 ; 24 : 5549 5557 . Crossref PubMed Google Scholar

16. Escola-Verge L , Rodriguez-Pardo D , Lora-Tamayo J , Morata L , Murillo O , Vilchez H , et al. Candida periprosthetic joint infection: A rare and difficult-to-treat infection . J Infect . 2018 ; 77 ( 2 ): 151 157 . Crossref PubMed Google Scholar

17. Kim J-K , Lee D-Y , Kang D-W , Ro D-H , Lee MC , Han H-S . Efficacy of antifungal-impregnated cement spacer against chronic fungal periprosthetic joint infections after total knee arthroplasty . Knee . 2018 ; 25 ( 4 ): 631 637 . Crossref PubMed Google Scholar

18. Brown TS , Petis SM , Osmon DR , Mabry TM , Berry DJ , Hanssen AD , et al. Periprosthetic Joint Infection With Fungal Pathogens . J Arthroplasty . 2018 ; 33 ( 8 ): 2605 2612 . Crossref PubMed Google Scholar

19. Theil C , Schmidt-Braekling T , Gosheger G , Idelevich EA , Moellenbeck B , Dieckmann R . Fungal prosthetic joint infection in total hip or knee arthroplasty: A retrospective single-centre study of 26 cases . Bone Joint J . 2019 ; 101-B ( 5 ): 589 595 . Crossref PubMed Google Scholar

20. Sidhu MS , Cooper G , Jenkins N , Jeys L , Parry M , Stevenson JD . Prosthetic fungal infections: poor prognosis with bacterial co-infection . Bone Joint J . 2019 ; 101-B (( 5 ): 582 588 . Crossref PubMed Google Scholar

21. Kuiper JW , van den Bekerom MP , van der Stappen J , Nolte PA , Colen S . 2-stage revision recommended for treatment of fungal hip and knee prosthetic joint infections . Acta Orthop . 2013 ; 84 ( 6 ): 517 523 . Crossref PubMed Google Scholar

22. Jakobs O , Schoof B , Klatte TO , Schmidl S , Fensky F , Guenther D , et al. Fungal periprosthetic joint infection in total knee arthroplasty: a systematic review . Orthop Rev (Pavia) . 2015 ; 7 ( 1 ): 5623 . Crossref PubMed Google Scholar

23. Belden K , Cao L , Chen J , Deng T , Fu J , Guan H . Hip and knee section, fungal periprosthetic joint infection . Diagnosis and Treatment: Proceedings of International Consensus on Orthopedic Infections J Arthroplasty . 2019 ; 34 ( 2S ): S91S387 . Crossref PubMed Google Scholar

24. Tarabichi M , Shohat N , Goswami K , Alvand A , Silibovsky R , Belden K , et al. Diagnosis of Periprosthetic Joint Infection: The Potential of Next-Generation Sequencing . J Bone Joint Surg Am . 2018 ; 100 ( 2 ): 147 154 . Crossref PubMed Google Scholar

25. Morawietz L , Classen R-A , Schröder JH , et al. Proposal for a histopathological consensus classification of the periprosthetic interface membrane . J Clin Pathol . 2006 ; 59 ( 6 ): 591 597 . Crossref PubMed Google Scholar

26. Schoof B , Jakobs O , Schmidl S , Klatte TO , Frommelt L , Gehrke T , et al. Fungal periprosthetic joint infection of the hip: a systematic review . Orthop Rev (Pavia) . 2015 ; 7 ( 1 ): 5748 . Crossref PubMed Google Scholar

27. Troendlin F , Frieler S , Hanusrichter Y , Yilmaz E , Schildhauer TA , Baecker H . Persistent Fistula for Treatment of a Failed Periprosthetic Joint Infection: Relic From the Past or a Viable Salvage Procedure . J Arthroplasty . 2020 ; 35 ( 2 ): 544 549 . Crossref PubMed Google Scholar

Author contributions

H. Baecker: Conceived and designed the study. Analyzed and interpreted the data, Provided administrative, technical, and material support, Wrote and edited the manuscript.

S. Frieler: Acquired, analyzed, and interpreted the data, Edited the manuscript.

J. Geßmann: Reviewed and edited the manuscript.

S. Pauly: Reviewed and edited the manuscript.

T. A. Schildhauer: Provided administrative support, Reviewed and edited the manuscript.

Y. Hanusrichter: Reviewed the literature, Conceived and designed the study, Acquired, analyzed, and interpreted the data, Performed the statistical analysis, Wrote the manuscript.

Funding statement

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

ICMJE COI statement

H. Baecker reports payment for lectures (including service on speakers bureaus) from Zimmer/Biomet, B.Braun/Aesculap, P.Brehm, and Infectopharm, all of which is unrelated to this work.

Open access funding

This manuscript has been self-funded for open access publication.

Ethical review statement

Ethical approval was obtained prior to the investigation from the local ethics committee (reference number 18 to 6538-BR). The study was approved by the German Clinical Trials Register (reference number DRKS00020409).

© 2021 Author(s) et al. This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/