Management and diagnostic guidelines for fungal diseases in infectious diseases and clinical microbiology: critical appraisal

Authors

  • S. Leroux,

    1. Department of Internal Medicine II, Universitätsklinikum, Julius Maximilian's University, Würzburg, Germany
    Search for more papers by this author
  • A. J. Ullmann

    Corresponding author
    1. Department of Internal Medicine II, Universitätsklinikum, Julius Maximilian's University, Würzburg, Germany
    • Corresponding author: A. J. Ullmann, Department of Internal Medicine II, Division of Infectious Diseases, Universitätsklinikum Julius Maximilian's University ,Oberdürrbacher Str. 6, 97080 Würzburg, Germany

      E-mail: andrew.ullmann@uni-wuerzburg.de

    Search for more papers by this author

Abstract

Invasive fungal infections (IFIs) are life-threatening conditions that require rapid diagnostic and optimal management to mitigate their high morbidity and mortality rate. They are also associated with a high economic burden, owing to prolonged hospitalization, the need for intensive supportive care, and the consumption of costly new antifungal agents. To address these issues, several international organizations have proposed guidelines for the management of IFIs. The consistency and reliability of these guidelines have rarely been assessed. This article is a review of the differences between the recommendations of the Infectious Diseases Society of America, the European Conference on Infection in Leukaemia, and the European Society of Clinical Microbiology and Infectious Diseases, and will focus on targeted treatment and diagnostic procedures. Although the conclusions of the three groups of experts are in many points similar we outlined some important differences in the methodology and conclusions of ESCMID. The use of these guidelines has the potential to enhance the management of fungal infections but is probably currently suboptimal.

Introduction

In recent years, invasive fungal infections (IFIs) have emerged as an important cause of life-threatening infection, especially in the context of immunosuppression, but also frequently in non-neutropenic patients, mainly in those requiring treatment in an intensive-care unit. Despite the growing body of evidence and knowledge in this field, the diagnosis and management of these complex infections remain challenging. International guidelines based on evidence-based medicine criteria that have been published in the past decade by three groups (including two scientific societies) will be the focus of this article: the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) [1-7], the Infectious Diseases Society of America (IDSA) [8, 9], and the European Conference on Infection in Leukaemia (ECIL) [10, 13]. The present review aims to find and analyse the discrepancies in the methodology (Table 1) and conclusions of these different guidelines for invasive candidiasis, aspergillosis, and mucormycosis. We will basically focus on diagnostic procedures and treatment recommendations, and exclude topics on combination therapies and prophylaxis. Children and neonates were also considered in guidelines provided by the ESCMID and the IDSA (partly summarized in Table 3), but not in those provided by the ECIL, but review of these would lead to this article exceeding the specified word limit. Therefore, the scope of this comparison was voluntarily restricted, and it is not an exhaustive assessment of each recommendation throughout the three guidelines.

Table 1. Comparison of the methodology and experts overlap
 ECILESCMID IDSA
  1. EBMT: European Group for Blood Marrow Transplantation, ECCM: European Confederation of Medical Mycology, EFISG: ESCMID Fungal Infection Study Group, ELN: European Leukemia Net, EORTC: European Organisation for Research and Treatment of Cancer, ESICM: European Society of Intensive Care Medicine, ICHS: Immunocompromised Host Society, QoE: quality of evidence, SPGC Standards and Practice Guidelines Committee, SoR: strength of recommendation.

PopulationAdult haematological malignancies patients and adult HSCT recipients, adult non-neutropenic patients (ICU)Paediatric patients, HIV patients, non-neutropenic adult patients (ICU), adult haematological malignancies and cancer patients and adult HSCT and SOT-recipientsPaediatric patients, HIV patients, non-neutropenic adult patients (ICU), adult haematological malignancies patients and adult HSCT and SOT-recipients
Scope

Diagnosis procedures

Antifungal prophylaxis

Invasive aspergillosis

Invasive Candidiasis

Mucormycosis

Diagnosis procedures

Anti-candida prophylaxis

Invasive candidiasis

Mucormycosis

Antifungal prophylaxis

Invasive, chronic and allergic forms of aspergillosis

Invasive candidiasis

Experts selection

All guidelines

EORTC, EBMT, ELN, ICHS

Candidiasis

EFISG, EBMT, ECCM, EORTC, ESICM

Mucormycosis

EFISG, ECCM

Candidiasis

IDSA, SPGC

Aspergillosis

Methodology not detailed

Evidence searchMedline, PubMed, Cochrane Library, Abstracts of ASH, ICAAC, ASCO, ESCMID and EBMT (limited to the 4 previous years)PubmedPubmed
Last year covered in search200920112009
Methodology and achievement of experts consensus

Organization committee defines topic and main questions

Working groups (3–6 experts) review the literature and prepare recommendations/updates

Meeting in ECIL with group consensus

EFISG defines clinical questions, intention of the recommendations and intervention

Working groups review the literature and prepare recommendations (note: literature database available to the whole panel via FTP-server)

2 Meetings in the ESCMID Conference on 2 consecutive years with group consensus

Note: Grading of the SoR and QoE in two separate evaluation

Presentation in workshop session in ESCMID, and incorporation of points of discussion

External peer review

Candidiasis guidelines:

Expert panel defines clinical questions

All experts review the literature and contribute to the preparation of the recommendation

Deliberations during

11 teleconferences

1 face-to-face meeting

External peer review

Approved by IDSA SPGC and Board of directors

Aspergillosis guideline:

Methodology not detailled

Expert overlap

With ESCMID : 15

With IDSA : 2

With ECIL : 15

With IDSA : 2

With ECIL: 2

With ESCMID: 2

Financial supportExperts meetings supported through educational grants from biomedical and pharmaceutical companiesESCMIDIDSA

Methods

The development of guidelines should be a standardized process, and the methodologies used by the three groups to develop these recommendations are quite similar. Nevertheless, there are some differences in term of objectives, scope, and editorial choices. In particular, it is important to note that the ECIL group restricted its recommendations to the group of patients with haematological diseases and to haematopoietic stem cell transplant (HSCT) recipients. There are also some important discrepancies in the methodology regarding the presentation of the strength of the recommendations (SoR) (Table 2). The IDSA uses a three-category system to rank its recommendations. This system was adopted by the ECIL group in its 2009 update of its guidelines, replacing the previous five-category system in order to harmonize the different international guidelines. The ESCMID proposed a four-category system in its recommendations on invasive candidiasis in 2012 (a further three ESCMID guidelines, which incorporate this ESCMID system of four categories, are in development or have been submitted for publication). There is a difference in the paradigm underlying this system in comparison with the IDSA. The latter mainly serves the purpose of ranking the quality of evidence (QoE), which leads to the recommendation's grade. The ESCMID system, on the other hand, aims at ranking the strength of the recommendations by the interpretation of an international expert group, based on their analysis of the literature. This difference seems minimal, but leads to very different conclusions in the case of some much debated issues (e.g. the application of amphotericin B deoxycholate). To further enhance the use of its guidelines in daily practice, the ESCMID group defines ‘intention’ for each of its recommendations (the intervention). According to the GRADE working group: ‘Recommendations to administer, or not administer, an intervention,’ should be ‘based on the trade-offs between benefits on the one hand, and risks, burden and, potentially, costs on the other.’ If benefits outweigh risks and burden, experts will recommend that clinicians offer a treatment to typical patients. The uncertainty associated with the trade-off between the benefits and risks and burdens will determine the SoR [14]. The ESCMID Candida guidelines followed this advice for grading a recommendation, as the QoE by itself is usually not sufficient. All three groups differentiate only three instead of the four categories proposed by GRADE to classify the QoE (high, moderate, low, or very low). The authors of this article think that three categories are sufficient to provide evidence, as at least the ESCMID QoE provides a more detailed subcategorization for level II [1].

Table 2. Presentation of the different grading systems for the strength of recommendation
Strength of recommendationECIL until 2009ESCMIDIDSA/ECIL since 2009
AStrong evidence for efficacy and substantial clinical benefit: strongly recommendedESCMID strongly supports a recommendation for useGood evidence to support a recommendation for or against use
BStrong or moderate evidence for efficacy, but only limited clinical benefit: generally recommendedESCMID moderately supports a recommendation for useModerate evidence to support a recommendation for or against use
CInsufficient evidence for efficacy, or efficacy does not outweigh possible adverse consequences (e.g. drug toxicity or interactions) or cost of chemoprophylaxis or alternative approaches: optionalESCMID marginally supports a recommendation for usePoor evidence to support a recommendation
DModerate evidence against efficacy or for adverse outcome: generally not recommendedESCMID supports a recommendation against useNA
EStrong evidence against efficacy or for adverse outcome: never recommendedNANA
  1. ECIL, European Conference on Infection in Leukaemia; ESCMID, European Society of Clinical Microbiology and Infectious Diseases; IDSA, Infectious Diseases Society of America; NA, not applicable.

  2. Added Index proposed by the ESCMID only for level II of evidence: r, meta-analysis or systematic review or randomized controlled trials; t, transferred evidence, i.e. results from different patient cohorts, or similar immune-status situation; h, comparator group is a historical control; u, uncontrolled trial; a, published abstract (presented at an international symposium or meeting).

Quality of evidence 
IEvidence from at least one well-executed randomized trial
IIEvidence from at least one well-designed clinical trial without randomization; cohort or case-controlled analytical studies (preferably from more than one centre; multiple time-series studies; or dramatic results from uncontrolled experiments
IIIEvidence from opinions of respected authorities based on clinical experience, descriptive studies, or reports from expert committees

Candidaemia/invasive candidiasis in haematological patients and HSCT recipients

The most updated guidelines for Candida infections were released by the IDSA in 2009 [8], the ECIL in 2011 [11], and the ESCMID in 2012 [1-6] (Table 3). For candidaemia in patients with haematological malignancies and HSCT recipients, the recommendations for first-line antifungal therapy are relatively consistent in the three guidelines, with echinocandins being the drugs of choice. The ECIL group gives a more moderate recommendation for the use of these agents, by grading them BII. Liposomal amphotericin B (LAmB) is recommended as a first-line therapy alternative with the same strength as echinocandins by the IDSA and the ECIL (respectively, AII and BII). Owing to its increased toxicity, the ESCMID proposes a weaker recommendation (BII), despite non-inferiority in comparison with micafungin. The other lipid formulations of amphotericin B are not considered in the IDSA guidelines, but the ECIL recommends their use with the same strength as for liposomal amphotericin B (BII). Owing to their toxicity profile and weaker database, the ESCMID proposes these agents for second-line treatment, with a CII recommendation (marginally supports a recommendation for use) for amphotericin B lipid complex (ABLC) and a CIII recommendation for amphotericin B colloidal dispersion (ABCD). Fluconazole is no longer considered to be a first-line therapy option. The IDSA considers its use to be possible in patients with mild to moderate disease and no prior exposure to azoles (BIII), whereas the ESCMID advises that it should be reserved as a step-down therapy option in case of infection with susceptible Candida species (CII). One of the main differences concerns the recommendation on the use of amphotericin B deoxycholate: the ECIL considers its use as an option (CIII), unless there is concomitant nephrotoxic medication (DIII) or renal impairment (EIII); the IDSA does not state a recommendation; and the ESCMID strongly recommends against its use in adults, because of its increased toxicity in comparison with the other, clearly better, therapy options (DII). There is also a difference in recommendations between the ESCMID and the two other groups concerning the follow-up strategy. The ESCMID recommends a switch to an echinocandin if the patients were receiving fluconazole or liposomal amphotericin B, e.g. for prophylaxis or empirical therapy. The IDSA considers it reasonable to pursue the same treatment if the patients are stable.

Table 3. Comparison of the recommendations on the management of candidaemia/invasive candidiasis
 ECILESCMIDIDSA
  1. 5-Fu: 5-fluorocytosine, ABCD: amphotericin B colloidal dispersion, ABLC: amphotericin B lipid complex, AmB-d: amphotericin B deoxycholate, CVC: central venous catheter, L-AmB: liposomal amphotericin B, NR: no recommendation, PICC: peripheral inserted central catheter, SoR: strength of recommendation.

  2. a

    D III, by concomitant nephrotoxic drug and E III by renal impairment.

  3. b

    Rather as step-down therapy.

  4. c

    Less critically ill patients, no azole exposure.

  5. d

    Not in severely ill patients or in patients with previous azole prophylaxis.

  6. e

    Not in patients with previous azole prophylaxis.

  7. f

    Not all experts agreed, SoR results from a majority vote.

  8. g

    May be better than echinocandins against C. parapsilosis.

  9. h

    Alternative if there is intolerance to or limited availability of other antifungals, transition to fluconazole if the patient is stable and sensible isolates recommended (A I).

  10. i

    Less critically ill patients, no recent azole exposure, step-down from echinocandins recommended if sensible species and patient stable (A II).

  11. j

    Consideration given to a loading-dose of 25 mg/kg.

Antifungal susceptibility testing

Recommended by lack of clinical response A II

Recommended to support a change in initial therapy B II

Recommended before a switch to oral azole A II

Recommended for all Candida strains isolated from blood and other deep sites (no SoR)

Recommended for superficial isolates in cases of no-response to treatment or relapses (no SoR)

Recommended for Candida glabrata (no SoR)

Recommended by failure to respond to initial antifungal therapy (no SoR)

Recommended if azole resistance is strongly suspected (no SoR)

Not recommended for Candida albicans (no SoR)

Therapeutic drug monitoringNR

Recommended for voriconazole and posaconazole and 5-FU (no SoR)

Recommended for echinocandins for patients with extra-corporeal membrane oxygenation (ECMO) (no SoR)

Recommended for voriconazole and itraconazole for patients with prolonged courses for deep-seated or refractory infections (no SoR)
Diagnosis procedures

Blood cultures: always recommended (no SoR)

Mannan/Anti-mannan:

For candidaemia: C II

For hepatosplenic candidiasis: B III

β-Glucan:

For haematological patients B II

Blood cultures: essential investigation (no SoR)

Mannan/anti-mannan:

For candidaemia: recommended II

 For invasive candidiasis: not recommended (no data)

For hepatosplenic candidiasis: recommended II

β-Glucan:

For candidaemia: recommended II

For invasive candidiasis: recommended II

NR
Initial targeted treatment in the haematological malignancies patients and HSCT-recipients

Caspofungin B II

Micafungin B II

Anidulafungin B II

L-AmB B II

ABLC B II

ABCD B II

Voriconazole B II

Fluconazole C III

AmB-d C IIIa

Itraconazole NR

Posaconazole NR

Caspofungin A IIt

Micafungin A IIt

Anidulafungin B IIt

L-AmB B IIt

Voriconazole C IIt

Fluconazole C IIta

ABLC C IIa

ABCD C III

AmB-d D IIt

Itraconazole D III

Posaconazole D III

Caspofungin A II

Micafungin A II

L-AmB A II

Anidulafungin A III

Fluconazole B IIb

AmB-d NR

ABLC NR

ABCD NR

Itraconazole NR

Posaconazole NR

Voriconazole NR

Initial targeted treatment in non-neutropenic patients

Caspofungin A I

Anidulafungin A I

Micafungin A I

L-AmB A I

AmB-d A Ia

Fluconazole A Ic

Voriconazole A Id

ABLC A II

ABCD A II

Itraconazole NR

Posaconazole NR

Caspofungin A I

Anidulafungin A I

Micafungin A I

L-AmB B I

Voriconazole B Ie

Fluconazole C Ie,f

ABLC C IIa

AmB-d D I

ABCD D IIa

Itraconazole D IIa

Posaconazole D III

Caspofungin A I

Anidulafungin A I

Micafungin A I

L-AmB A Ig

AmB-d A I

Fluconazole A Ih

Voriconazole A I

Itraconazole NR

Posaconazole NR

ABLC NR

ABCD NR

Initial targeted treatment in neonatesNR

AmB-d 1 mg/kg/day B II

L-AmB 2.5–7 mg/kg/day B II

Fluconazole 12 mg/kg/day B II

Micafungin 4–10 mg/kg/day B II

Caspofungin 25 mg/m²/day C II

ABLC C II

AmB-d &mg/kg/day A II

L-AmB 3–5 mg/kg/day B II

Fluconazole 12 mg/kg/day B II

Echinocandins B III

ABLC NR

Combination therapyNR

AmB-d + fluconazole D I

AmB-d + 5-Fu D II

Efungumab + lipid-associated amphotericin B D II

NR

Intravenous catheter removal

Non-haematological patients A II

Haematological patients B III

Candida parapsilosis infection A II

Non-neutropenic patients A IIr

Haematological patients A IIu

Non-neutropenic patients A II

Neutropenic patients: B III

Treatment duration

Follow up blood cultures: NR

Non-neutropenic: 14 days after the last positive blood culture and resolution of signs and symptoms B III

Neutropenic patients: 14 days after the last positive blood culture and resolution of signs and symptoms and resolution of neutropenia C III

Take at least one blood culture per day until negative B III

Non-neutropenic patients: treat 14 days after the end of candidaemia B II

Neutropenic patients: treat at least 14 days after last positive blood culture, if neutropenia persist appropriate duration not defined (no SoR)

Blood cultures should be performed daily or every other day (no SoR)

Non-neutropenic 14 days after clearance of Candida from the bloodstream A III

Neutropenic patients: 14 days after clearance of Candida from the bloodstream and resolution of the neutropenia (no SoR)

Follow up

Importance of an active search for dissemination of infection in leukemic patients following neutrophil recovery (ocular fundus + abdominal imaging) (no SoR)

Step down therapy: NR

Transoesophageal echocardiography B IIa

Fundoscopy B II

If CVC, PICC or intravascular devices, search for thrombus B III

Step-down to fluconazole after echinocandin after 10 days of IV, if species is susceptible, patient tolerates PO and patient is stable B II

Fundoscopy (no SoR)

Transition from an echinocandin to fluconazole for non-neutropenic patients who have isolates likely to be susceptible to fluconazole A I

The management of central venous catheters in the context of candidaemia is less straightforward in this population of patients than in non-neutropenic patients, owing to the increased morbidity associated with catheter replacement. The IDSA guidelines only state that removal should be considered for patients who have persistent candidaemia (BIII); the ECIL gives a similar recommendation (BIII), unless Candida parapsilosis is involved, in which case the authors consider removal to be mandatory (AII), this species being more frequently associated with catheter-associated infections. The ESCMID supports more clearly early catheter removal (AII) if feasible, and the use of echinocandins (or liposomal amphotericin B) in cases of clinically necessary catheter retention (CII). This is also the only group that proposes recommendations for the use of colony-stimulating factors (granulocyte colony-stimulating factor), with a weak recommendation (CIII), and granulocyte infusion, which might be an option in desperate cases (CIII).

Candidaemia/invasive candidiasis in non-neutropenic patients

In its guideline for patients without neutropenia, the ESCMID provides a strong recommendation (AI) for the use of all echinocandins in comparison with the lipid formulation of amphotericin B (BI) and voriconazole (BI), whereas the ECIL and the IDSA provide a strong recommendation for the use of these alternatives. A main difference is that the ESCMID guideline does not consider fluconazole to be an agent for first-line therapy any more (CI), but rather as an oral step-down therapy option in cases of good response to treatment, owing to its limited spectrum (concerns regarding resistance) and inferiority in a randomized study vs. anidulafungin [12, 15]. The ESCMID document states that fluconazole might be an option for C. parapsilosis infections because of uncertainty about MICs.

Diagnostic procedures

Only the ECIL in its 2009 update and the ESCMID in 2012 made recommendations for the diagnosis of invasive candidiasis and candidaemia [2, 11, 13]. Conventional diagnostic procedures, such as microscopic examination, sample and blood culture, and identification to the species level, are standardized and form part of daily clinical practice, but important details were summarized. There was a need for recommendations concerning the new non-culture-based techniques. Regarding its other recommendations, the ECIL group limited its analysis to the setting of patients with haematological malignancies and HSCT recipients. Nevertheless, their conclusions are very similar. The ESCMID moderately recommended the use of the β-d-glucan assay in the diagnosis of invasive candidiasis, candidaemia, and chronic disseminated candidiasis, and the use of mannan/anti-mannan assays in the diagnosis of the two latter conditions, but not for invasive candidiasis, for both neutropenic and intensive-care unit patients. The ECIL also moderately supports these tests, with a lower strength of recommendation for the mannan/anti-mannan assay in candidaemia (CII). Moreover, the ESCMID suggests the use of serial determination for the β-d-glucan assay, and emphasizes its utility for ruling out infections.

Invasive aspergillosis

Guidelines on the management of invasive aspergillosis have profited from data from randomized trials with a high quality of evidence. Thus, the two international guidelines proposed by the IDSA in 2000 and updated in 2009 (9) and those of the ECIL group published in 2005 and updated three times up to 2009 are very similar for neutropenic patients and HSCT recipients [10]. The ESCMID is in the process of developing a guideline on this topic.

Ubiquitously, voriconazole 6 mg/kg twice-daily and then 4 mg/kg twice-daily is the treatment of choice, supported by several randomized studies, with a strong recommendation (AI) in both guidelines. The ECIL specifies that therapy should begin with the intravenous form of voriconazole, owing to insufficient pharmacokinetic data supporting oral therapy (oral initiation graded CIII). The IDSA recommends beginning therapy with the parenteral formulation for seriously ill patients (AIII). Neither gives any recommendation for the duration of the parenteral administration or the overall duration of the therapy. The recommendations for diagnosis strategy, indication for surgery, adjuvant treatment with colony-stimulating factors, management of immunosuppression and corticosteroid therapy also appear to be very similar.

Mucormycosis

Despite mucormycosis being the third most common type of IFI, there has been, until recently, a lack of recommendations on diagnosis and the best management. The data on the management of these infections are scarce, and consist mainly of case series and experimental studies. On the basis of this literature, the ECIL made recommendations in 2009 [12] and the ESCMID in 2013 [7]. Their conclusions are very similar concerning diagnostic procedures, first-line and salvage therapy, and adjunctive treatment, in particular the recommendation against the use of the iron chelator deferasirox and the scarcity of data in favour of hyperbaric oxygen, and both guidelines emphasize the importance of surgery in synergy with antifungal therapy and the control of underlying conditions (e.g. neutropenia and diabetes). It is important to note here that there is some overlap of authors between ECIL and ESCMID guideline but basically for patients with haematological malignancies. Only one point differs significantly between the two documents: the ECIL considers that posaconazole cannot be recommended for first-line therapy (CIII), whereas the ESCMID considers posaconazole 200 mg four times daily or 400 mg twice-daily as a therapy option, albeit with a weaker strength of recommendation than for liposomal amphotericin B (BII). This difference in the strength of recommendations is explained mainly by two studies assessing the use of posaconazole as first-line therapy, which were published after the ECIL guidelines [16, 17].

Finally, we note that the ESCMID discourages the use of amphotericin B deoxycholate, which is consistent with the discussions that took place regarding the management of candidiasis in adults [1, 7]. Some items remain unresolved in both guidelines, in particular the lack of a clear duration or therapeutic endpoint, but this is not too surprising since published data on this matter is rare

Discussion

The IDSA, ECIL and ESCMID aim at providing evidence-based international guidelines for the diagnosis and treatment of IFIs. The clinical recommendations from the expert panels from these three organizations (two scientific societies [ESCMID and IDSA] and one working group [ECIL]) are, in important respects, consistent, and the fact that their conclusions are similar enhances their reliability. Although the limited number of experts and the overlap of authors temper this observation for the mucormycosis guidelines. A major difference is the position of the ESCMID experts on the use of amphotericin B deoxycholate in adults. Their final decision is based on a balance between the accrued toxicity without additional benefits and the cost incurred by the use of alternative agents that they feel to be acceptable, at least in Europe. They further emphasize that the cost-effectiveness of the different therapeutic and diagnostic strategies are difficult to assess, because of the diversity of reimbursement systems in Europe, and that their guidelines should also be easily adapted to local clinical practice, with due respect to the local fungal epidemiology and economic considerations. This also applies to the strong recommendation on the use of echinocandins over fluconazole in the first-line therapy of candidaemia.

Another important issue is that the grading systems are not comparable. The ECIL chose to modify its grading system during the last update of its guideline, which is now identical to the IDSA's. This makes comparison between the most recent recommendations of the ECIL and the IDSA easier, but may make them more difficult to understand and difficult to use in daily clinical practice. On the other hand, the ESCMID tries to emphasize this aspect in the composition of their guideline by defining clear patient groups, and precise interventions and intentions for each recommendation, and using a more definite grading system. The ESCMID “strength of recommendation” system is the only one proposing the need for further research if a grade C is applied to further improve the scientific grounds for a given intervention [1].

Some studies have tried to assess adherence to the international recommendations and its consequences. In their retrospective monocentre study, Nivoix et al. [18] evaluated the adherence to the recommendations from the ECIL and the IDSA for 199 antifungal treatments in 133 adults, and found a low rate of appropriate prescription (34%). The 12-week survival rate was lower in the group of patients receiving inappropriate or debatable treatment (70% vs. 81%), but not significantly (p 0.24). The very low rate of adherence to the guidelines in this study could be attributable to the very stringent criteria defining ‘appropriate’ treatment. Another study, from Patel et al. [19], assessed not only the choice and dosage of antifungal therapy but also the global adequacy of the management of 199 patients with candidaemia with regard to the recommendations from the IDSA. The therapy was consistent with the guidelines in 76% of patients, and deviation from these recommendations was associated with increased mortality (24% vs. 57%, p 0.003). In a retrospective multicentre study, Pagano et al. [20] observed 55% adherence to IDSA guidelines and 28% adherence to ECIL guidelines in 136 patients with acute leukaemia and proven or probable invasive aspergillosis. The first-line therapy response rate was significantly better in the group with adherence to the IDSA (71%) and the ECIL (84%) guidelines than in the remaining patients (respectively, 59% and 62%), although no difference could be seen regarding the rate of survival at 120 days. It is important to note that all of these studies took only the IDSA and ECIL guidelines into account. The ESCMID guidelines have not yet been assessed.

These results support the utility of international updated guidelines for the management of fungal infections but, at the same time, emphasize the low compliance rate and the need to enhance their acceptance and usage in daily practice. These results support the utility of international updated guidelines for the management of fungal infections but in the same time emphasize the low compliance rate and the need to enhance their acceptance and usage in the daily practice. The reasons underlying the lack of compliance to guidelines are multifactorial [21] but the adaptation of international structural guidelines to the local level, with integration of prescribers into the process, is likely to enhance their diffusion and acceptance rate. Another argument to the adaptation to the local setting is the important geographical variations in the epidemiology. In this, the role of the specialist in infectious disease seems to be pivotal and the inclusion of antifungal stewardship in the antimicrobial stewardship programs the keys to an optimisation of the management of fungal diseases [22]. To this respect, the need of several international guidelines in different areas of the world might not be that questionable.

In the near future further collaboration might fill the gap for streamlined recommendations. It could resolve the problem of the most debated and inconsistent points and harmonize the grading system for the strength of recommendation and quality of evidence, that are issues in the adherence due to the difficulty of comparing the current guidelines. The GRADE working group could be a possible solution. When focussing on those differences a real international guideline comes within the reach of a possibility. The main obstacles in the development of such a guideline remain, as stated in the introduction of ESCMID recommendations, the need to create a useful and comprehensive framework that could be utilized in local modified operating procedures by physicians and the difficulty of balancing the economic considerations and complexity of national reimbursement system with recommendations based on sound evidences from controlled clinical trials. While covering all aspects and opinions an international guideline for all will remain a challenge.

Transparency Declaration

S. Leroux has no potential conflicts of interest. A. J. Ullmann has received research grants from MSD (Schering Plough), and is/was an advisor or received lecture honoraria from Astellas, Aicuris, Basilea, Gilead, MSD, and Pfizer.

Ancillary