Accepted Manuscript

Management of Invasive Candidiasis and Candidemia in critically ill adults – Expertopinion of the European Society of Anaesthesia (ESA) Intensive Care scientificsubcommittee

Ruth-Aoibheann OLeary, Sharon Einav, Marc Leone, Krisztina Madách, ClaudeMartin, Ignacio Martin-Loeches

PII: S0195-6701(17)30647-3

DOI: 10.1016/j.jhin.2017.11.020

Reference: YJHIN 5290

To appear in: Journal of Hospital Infection

Received Date: 14 September 2017

Accepted Date: 29 November 2017

Please cite this article as: OLeary R-A, Einav S, Leone M, Madách K, Martin C, Martin-Loeches I,Management of Invasive Candidiasis and Candidemia in critically ill adults – Expert opinion of theEuropean Society of Anaesthesia (ESA) Intensive Care scientific subcommittee, Journal of HospitalInfection (2018), doi: 10.1016/j.jhin.2017.11.020.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

Management of Invasive Candidiasis and Candidemia in critically ill adults – Expert opinion of 1 the European Society of Anaesthesia (ESA) Intensive Care scientific subcommittee 2 3 Ruth-Aoibheann OLeary (1) Sharon Einav (2). Marc Leone (3), Krisztina Madách (4), Claude 4 Martin (3), Ignacio Martin-Loeches (1), 5

6

Institutional Affiliations: 7

1. Multidisciplinary Intensive Care, St James's University Hospital, Department of Clinical 8 Medicine, Trinity College, Welcome Trust-HRB Clinical Research Facility, St James Hospital, 9 Dublin, Ireland. 10

2. General Intensive Care unit, Shaare Zedek Medical Centre and Hebrew University Faculty 11 of Medicine, Jerusalem, Israel. 12

3 Aix Marseille University, Anaesthesia and intensive care unit and trauma centre, Nord 13 Hospital, Assistance Publique Hôpitaux de Marseille, APHM, Marseille, France 14

15 4. Department of Anaesthesiology and Intensive Therapy, Semmelweis University, Budapest, 16 Hungary 17

18 Word Count: 5766 19

20

Assistance with the study: none. 21

Financial support and sponsorship: none. 22

Conflict of interest: none. 23

24

25

26 27 28 29 Correspondence to: 30 Ignacio Martin-Loeches. St James's University Hospital, Dublin 8, Ireland. Phone: (01) 410 31 3000 32 E-mail: drmartinloeches@gmail.com 33 34

35

Keywords: candida, intensive care, critical care, invasive candidiasis, management, diagnosis 36

37

38

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

ABSTRACT 1

OBJECTIVE: The global burden of invasive fungal disease is increasing. Candida albicans 2

remains the leading cause of fungal blood stream infections although non-albicans Candidal 3

infections are emerging. Areas of controversy regarding diagnosis and management are 4

hampering our ability to respond effectively to this evolving threat. The purpose of this 5

narrative review is to address current controversies and provide recommendations to 6

supplement guidelines. 7

8

DIAGNOSIS OF INVASIVE CANDIDIASIS: Diagnosis of invasive candidiasis requires a 9

combination of diagnostic tests and patient risk factors. Beta-D Glucan and Candida albicans 10

germ tube antibody are both used as biomarkers adjuncts to diagnosis although direct 11

culture remains the gold standard. Scoring systems are available to help distinguish 12

between colonisation and invasive disease. 13

14

TREATMENT OF INVASIVE CANDIDIASIS: Echinocandins are recommended as first line 15

therapy in candidemia with de-escalation to fluconazole when clinical stability is achieved. 16

Empirical therapy is highly recommended in high-risk patients but a more targeted pre-17

emptive approach is now being favoured. The evidence for prophylactic therapy remains 18

weak. 19

20

SUMMARY: Mortality attributable to invasive candidiasis may be as high as 70%. Prompt 21

diagnosis and treatment, in conjunction with source control, are the key to improving 22

outcomes. 23

24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

3

INTRODUCTION 1

The global burden of Invasive Fungal Disease (IFD) is increasing. There is an urgent need to 2

implement antifungal stewardship programs in critically ill patients (1) due to an overuse of 3

antifungal agents in up to 74% of cases (2). There are many areas of controversy regarding 4

diagnosis and management of invasive Candida infection. Firstly, isolation of Candida from 5

microbiological samples does not clearly correlate with the burden of the disease. Therefore, 6

with the exception of patients with positive blood cultures, there is no direct evidence 7

showing that antifungal administration reduces mortality rates. In intra-abdominal 8

infections an association has been found between the presence of Candida species in 9

sampling and subsequent mortality rates (3,4) but the use of antifungals did not affect 10

patient outcome. Secondly, the timing and choice of antifungal therapy remains a matter 11

for debate, with convincing evidence that delayed treatment of candidemia is an important 12

determinant of patient outcome. However, septic shock attributed to Candida infection and 13

its determinants of outcome remains an evolving area of research (4,5). We have published 14

this narrative review with the aim of addressing these controversies and providing some 15

recommendations to supplement existing guidelines. 16

17

EPIDEMIOLOGY AND RISK FACTORS OF INVASIVE FUNGAL DISEASE 18

Traditionally, IFD has been associated with immunocompromised states, particularly 19

involving patients with malignancies, chronic inflammatory diseases and chronic 20

immunosuppressive conditions. Thus, the number of patients at risk is rising in intensive 21

care units (ICUs) (6). Candida species are the most common fungal pathogens associated 22

with health-care infections of the bloodstream in ICU (7). The leading cause of fungal blood 23

stream infections among patients admitted to an ICU remains Candida albicans, but non-24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

4

albicans Candida infections are also an emerging problem (8). Attributable mortality 1

associated with invasive candidiasis (IC) remains unacceptably high, ranging 5-71% 2

depending on clinical manifestations and timing of therapy administered (9,10). 3

4

Candida species are part of the human microbiota, colonising the skin, the gastrointestinal 5

and genitourinary tracts (11). IC is generally caused by a change in the concentration and 6

distribution of normal flora combined with an alteration of the barrier function of the 7

mucous membranes. Critical illness and critical care therapies lead to dysbiosis and 8

disruption of the normal barrier mechanisms, resulting in translocation of flora, including 9

Candida (12). Prolonged ICU stay and the use of broad-spectrum antibiotics are associated 10

with IC. Other risk factors include indwelling central venous catheters (CVC), particularly 11

those used for total parenteral nutrition; haemodialysis; recent major gastrointestinal 12

surgery; necrotizing pancreatitis; diabetes and any pre-existing cause for 13

immunosuppression (13). Moreover, the pathogenicity of Candida is enhanced by its ability 14

to adhere to prosthetic surfaces and form biofilms (14). The formation of biofilms plays an 15

important role in antifungal, especially azole, resistance (15). 16

17

The pathogenesis of IC involves all elements of the immune system. Lymphocytes are 18

essential for cell-mediated immunity to Candida and for the prevention of increased 19

colonization. Neutrophils and monocytes interfere with the reproduction of Candida by 20

damaging pseudohyphae and blastospores. Thus, patients with neutrophil dysfunction or 21

leukopenia are at risk for IC. This may lead to acute disseminated candidiasis (ADC), 22

characterised by a diffuse haemorrhagic and papular rash and small vessel vasculitis, with 23

multiple organ involvement (e.g. lungs, GI tract, liver, spleen, kidneys). ADC is associated 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

5

with a high mortality and is often seen at autopsy (16). Patients recovering from 1

chemotherapy-induced neutropaenia may present with chronic disseminated candidiasis. 2

The reappearance of functional neutrophils leads to the formation of multiple parenchymal 3

lesions in the splanchnic organs, detectable on imaging. These patients often present with 4

right upper quadrant abdominal pain, pyrexia and elevated alkaline phosphatase and 5

biopsies are usually needed for diagnosis (17). Finally, complement and immunoglobulins 6

play an important role in opsonisation and intracellular killing of pathogens so deficiencies in 7

these elements may lead to refractory infection (18). 8

9

Table 1 summarizes the major risk factors for IC. 10

11

DIAGNOSIS OF INVASIVE CANDIDIASIS 12

The diagnosis of IC involves the interpretation of fungal diagnostic tests in the context of 13

patient risk factors and clinical and radiological assessment (Table 2). The clinical 14

manifestations of IC are largely dependent on the site involved. Candidemia is the most 15

commonly diagnosed form of IC. However, IC has a wide range of manifestations. 16

17

Ocular involvement, manifesting as endophthalmitis, chorioretinitis or vitritis can occur in up 18

to 20% of cases, so ophthalmological examination is recommended in IC in ICU patients (19). 19

In neutropenic patients, a further ophthalmological examination should be performed after 20

neutrophil recovery(20). Involvement of bones and vertebral discs, in the form of 21

osteomyelitis and discitis, is an increasing problem mainly associated with untreated IC (21). 22

23

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

6

The incidence of Candida endocarditis has risen in parallel with the general increase in 1

Candida infections. Candida endocarditis is characterised by persistent pyrexia and 2

candidemia in blood cultures, new heart murmurs, often with worsening cardiac function 3

and embolic phenomena. It may be associated with recent cardiac surgery, prolonged CVC 4

use, chemotherapy or intravenous drug abuse (22). Critical care patients have a high risk of 5

central nervous system (CNS) involvement, often manifesting as meningitis, as a 6

complication of candidemia or as a complication of IC in presence of CNS prosthetic devices 7

(23). This occurs more frequently in severely immunosuppressed hosts. 8

9

The gold standard IC diagnostic test remains morphologic/phenotypic assessment, requiring 10

culture from sterile sites such as blood, pleural or peritoneal fluid or histopathology. Direct 11

tissue samples are desirable but it may not be feasible to obtain these in many clinical 12

situations. Pure cultures remain essential for antifungal susceptibility testing. This gives an 13

estimate of the in vitro potency of an antifungal agent against the cultured pathogen and is 14

very valuable as elevated minimum inhibitory concentration (MIC) is associated with 15

breakthrough infection and worse outcome (24). Antifungal susceptibility testing is 16

particularly relevant now that fluconazole resistance is detected in up to 50% of non-17

albicans Candida infections (25) . It also plays a role in de-escalation strategies (e.g. from 18

echinocandins to fluconazole), thus reducing treatment costs and antifungal pressure. 19

20

Several laboratory tests have been proposed as adjuncts for diagnosing the presence of IC 21

(Table 2). Assays of surrogate markers (i.e. antigens or antibodies related to fungal wall 22

components, fungal-related nucleic acids) have also been proposed to improve early 23

diagnosis of IC. Martin-Mazuelos et al conducted a prospective study in critically ill patients 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

7

and found that (1,3)-ß-D-glucan (BDG) levels were higher in patients with IC and high-grade 1

candida colonization(26). Two consecutive BDG levels ≥ 80 pg/mL allowed discrimination 2

between IC and high-grade colonization, but the area under the receiver operating curve for 3

both BDG and Candida albicans germ tube antibody (CAGTA) was low (0.67 and 0.55 4

respectively), indicating that this test cannot be used without supportive clinical correlates 5

(24). Similarly, more commonly used markers such as mannan and anti-mannan could be 6

useful as diagnostic tests. It is important to highlight that mannan tests have a very low 7

specificity (27). In the future, biomarker kinetics may be used to determine the presence of 8

fungal infection with greater precision, thereby ensuring appropriate recommendations for 9

early antifungal treatment. 10

11

Molecular methods, which are being gradually implemented in fungal diagnostics, may open 12

up a new dimension in the near future. However, at present they still have serious 13

limitations. PCR assays lack standardization regarding DNA isolation and choice of primers, 14

and contamination with fungi ubiquitous in the environment can happen at any stage of 15

fungal PCR. Due to these issues no single test was found to be accurate enough to be 16

incorporated in guidelines (28). Other molecular methods such as matrix-assisted laser 17

desorption/ionization (MALDI) time-of-flight mass spectrometer (TOF) - a postculture 18

technique requiring growth of the organism using matrix assisted laser desorption ionization 19

time-of-flight mass spectrometry and fluorescent in situ hybridization (FISH) performed on 20

positive blood cultures - may shorten species identification (29,30). 21

22

One of the major difficulties in the diagnostics of IC is differentiating between infection, 23

colonization and contamination. Two methods can be helpful in this situation: 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

8

1) Definitions created by the European Organization for Research and Treatment of 1

Cancer/Mycoses Study Group (EORTC/MSG) for invasive fungal diseases (IFD), which 2

were created for research purposes. They apply to immunocompromised patients 3

but not necessarily to critically ill patients (31) 4

2) Scores assessing the severity of Candida colonization, which correlate with the 5

occurrence of IC (32), and are meant to assess immunocompetent patients. 6

The EORTC/MSG definitions create a usable framework for defining proven, probable, 7

possible and unlikely presence of IFDs (31). These diagnostic criteria have proven to be 8

useful in research and practice in severely immunocompromised patients but not in 9

intensive care unit (ICU) patients. According to these definitions the presence of IFD can be 10

considered proven when the pathogen is found in normally sterile tissue/blood – either by 11

microscopic analysis or by positive cultures. Cases of probable IFD require a host factor, 12

clinical features including imaging findings, and mycological evidence – either direct tests 13

like cytology, microscopy, culture; or indirect tests like mannan or β-D-glucan. IFD is 14

possible with appropriate host factors and with sufficient clinical evidence but without 15

mycological support. 16

17

The rate of invasive fungal infections described in the literature is highly variable - often 18

because these definitions of IFD are still not being adhered to. Estimates may thus vary 19

widely even within regions of the same country and over time in the same region (33,34). 20

For example, the Extended Prevalence of Infection in Intensive Care II (EPIC II) study 21

reported a 19% prevalence of fungal infection based on positive isolates taken from ICU 22

patients. The authors noted that the presence of Candida in any isolate type was not 23

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

9

associated with in-hospital mortality (35). However, this single-day point-prevalence study 1

did not differentiate between true pathogens and colonization. Candida species comprise 2

part of the normal human flora and may typically found in the oral cavity, the 3

gastrointestinal and genitourinary tract. They also often colonize the lower airways in 4

critically ill patients, but do not necessarily cause infection. Candida pneumonia is very rare 5

and limited to seriously immunocompromised patients so Candida isolated from respiratory 6

secretions should not be considered pathogenic, unless histopathology can confirm the 7

presence of IC. In view of these inconsistencies it is unsurprising that the EPIC II study 8

contrasted with prior retrospective studies, which associated delayed administration of 9

antifungal therapy in patients with candidaemia with increased mortality (36). This 10

discrepancy exemplifies how lack of standardisation in the criteria used to diagnose fungal 11

infections in many studies (for example inclusion of both colonization and true infection), 12

remains a major setback to reaching plausible research conclusions on this topic. Clinical 13

guidelines acknowledge that only a small proportion of ICU patients will actually develop 14

true IC (37). 15

16

Colonisation with Candida is associated with the occurrence of invasive candidiasis (32). 17

Therefore clinical risk-assessment tools have been proposed by several authors to determine 18

the likelihood of invasive candidiasis compared with colonization in non-neutropaenic 19

critically ill patients. These include the Candida Colonization Index (CI), the Corrected 20

Colonization Index (CCI) and the Candida Score (CS). The CI is the ratio of the number of 21

distinct non-blood body sites colonized by Candida species to the total number of body sites 22

cultured. Samples may include oropharyngeal swabs, tracheal secretions, gastric fluid, 23

perineal swabs, stool samples, urine samples, surgical wound swabs, abdominal drain fluids 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

10

and catheter insertion sites. Only strains of Candida species with identical electrophoretic 1

karyotypes are included (38). The CS was developed in 2006 is comprised of the sum of the 2

following risk factors associated with invasive candidiasis: use of total parenteral nutrition 3

(TPN) (1 point), surgical patient (1 point), multifocal Candida colonization (1 point) and 4

current presence of severe sepsis (2 points). When using this score, a value ≥ 2.5 suggests 5

antifungal treatment might be required. The sensitivity of this score for proven (i.e. 6

microscopically seen or culture positive) infection rates is >0.8, with a specificity of 0.74 and 7

there have been studies demonstrating that the use of the CS may promote timely initiation 8

of antifungal treatment in cases of true fungal infection (39). In further studies, such as the 9

validation of the CS from Leon et al in 2009 (40) and a more recent study from Leroy et al in 10

2011 (41), the diagnostic CS value used was 3 and we consider that this threshold would 11

have a better predictive value. Different Candida colonization scores have different 12

predictive values for the likelihood of developing an invasive fungal infection, largely 13

dependent on the variables included in each score. At present, careful consideration of past 14

medical history and fungal diagnostic tests combined with clinical and radiological 15

assessment and clinical scores remain the best method for diagnosing the presence of IC in 16

our ICU patients. 17

18

TREATMENT OF INVASIVE CANDIDIASIS 19

In the critical care setting, the treatment of infection due to Candida remains a challenge. 20

Systemic antifungal therapy is widely used, in up to 7.5% of intensive care patients (42), 21

despite the fact that two-thirds of these patients have no documented IFD. Prophylactic and 22

empirical approaches to therapy may lead to overtreatment of patients with an increase in 23

resistant organisms whereas delay in treatment may increase mortality (5). Thus, a balance 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

11

must be found between timely and targeted treatment. The recommendations below are 1

based on clinical guidelines (20,43) and expert opinion. 2

3

Increases in resistance patterns locally and internationally have influenced the 4

recommended initial therapy for systemic candidiasis (see Table 3, 4). Echinocandins are 5

strongly recommended as the initial therapy for candidemia (20,43). Patients may 6

subsequently be switched to fluconazole after 5-7 days if clinical stability is achieved and 7

blood cultures remain negative (20). In view of this, blood cultures should be repeated 8

regularly until they are negative. Blood cultures could be repeated with a 1-day interval 9

rather than every day in the presence of fever. If a central venous catheter is suspected to 10

be the source of candidemia it is strongly recommended that the device should be removed 11

(20,43). Treatment should be continued for two weeks after blood cultures are negative and 12

clinical symptoms are resolved. Combination therapy is not recommended for candidiasis. 13

14

Empirical antifungal therapy is most broadly defined as administration of antifungals in 15

patients with refractory pyrexia and other risk factors for IFD (44). Empirical antifungal 16

therapy is strongly recommended in cases of suspected invasive candidiasis in non-17

neutropaenic patients in the presence of clinical risk factors (table 1)(20) and positive 18

surrogate markers, such as beta-D glucan, and pyrexia with no other cause. In patients with 19

septic shock and suspected candidiasis, empirical treatment should be started as soon as 20

possible and continued for the same duration as recommended for candidemia. If there is 21

no response to empirical treatment after 4-5 days and no evidence of invasive candidiasis, 22

clinicians should consider stopping treatment. 23

24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

12

Current guidelines recommend empiric antifungal therapy(20), however this often fails to 1

provide any benefit in ICU patients (45,46) and may result in significant over-treatment. 2

Given the low efficacy of this approach, clinicians should rely more on a combination of risk 3

factors and biomarkers. (1→3)-β-D-glucan (BG) and mannan, both fungal cell wall 4

components, are candidate biomarkers for fungal infection. On the one hand mannan is a 5

highly immunogenic antigen polysaccharide, immunologically more active then β-glucan and 6

positive results may be obtained 2-15 days before positive blood cultures. On the other 7

hand, (1→3)-β-D-glucan is an exo-antigen that can be detected in serum, BAL or CSF and 8

present in most Aspergillus (47). Thus, both may be of use in guiding therapy but mannan is 9

more relevant for Candidal infections. Of note, combined mannan antigen and antimannan 10

antibody testing is more sensitive than either test by itself (48). This more targeted 11

approach is referred to as the pre-emptive approach (49). 12

13

The pre-emptive approach is defined as administration of antifungals in patients with a 14

prolonged ICU stay; previous broad-spectrum antibiotic therapy; other clinical risk factors 15

and microbiological evidence of Candida infection, including multifocal colonisation or a 16

positive β-D glucan (50). The knowledge of epidemiological data, clinical risk factors and, 17

above all, the analysis of local epidemiology of candidemia in ICU allows one to determine 18

whether a patient is at low or high risk of developing this infection. There are currently no 19

validated definitions to define high versus low risk other than clinical assessment based on 20

different criteria. From expert opinion expressed in reviews and guidelines (20), therefore, a 21

high rate of invasive candidiasis is defined when there is >10% risk, as compared with the 22

normal rates of 1-2%. The pre-emptive approach is an attractive approach as it still allows 23

prompt treatment of high-risk patients while limiting the use of antifungals in low-risk 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

13

patients. However, scientific evidence is lacking to support its wider introduction so 1

prospective trials are needed for validation (45,51). When pre-emptive therapy is 2

considered, β-D glucan concentration should be ≥ 60 pg/mL or Candida score should be ≥3 3

(52,53). 4

5

Prophylactic therapy is defined as the administration of antifungals according to protocol in 6

patients with no definite evidence of infection. There is only moderate-quality evidence 7

giving a weak recommendation to use fluconazole or an echinocandin to prevent invasive 8

candidiasis in ICU patients. Potential benefits in terms of reduction in candidaemia rates are 9

balanced by risks of toxicity and increased resistance patterns. A Cochrane analysis noted 10

that the number needed to treat varied from 9 in high-risk patients to 188 in low-risk 11

patients (45). In view of this, antifungal prophylaxis is only recommended in high risk 12

patients in ICUs with a high incidence of invasive candidiasis (20,37). 13

14

With regard to suspected abdominal candidiasis, empirical therapy (see table 3) should be 15

considered for patients with intra-abdominal infection and significant clinical risk factors, 16

such as recent abdominal surgery, anastomotic leaks and necrotising pancreatitis (20). The 17

presence of Candida on direct examination of peritoneal fluid in the setting of 18

intraabdominal infection is significant and is associated with increased mortality (3). In the 19

presence of three of the four following criteria - haemodynamic failure, upper digestive tract 20

perforation, female gender, and antibiotic treatment within the previous 48 h – the 21

probability of identifying Candida in the peritoneal fluid is 71%, encouraging the empiric use 22

of antifungals. However, the scientific evidence is weak with studies of poor quality (54). 23

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

14

1

The interpretation of the significance of Candida isolates from non-sterile sites can be 2

difficult. For most, if not all, non-immunocompromised patients growth of Candida from 3

respiratory secretions indicates colonisation rather than pneumonia due to Candida. As a 4

result, the decision to treat should never be made on the basis of respiratory samples alone 5

(55). Similarly, in cases of asymptomatic candiduria treatment is not recommended unless 6

the patient is neutropenic or undergoing urological procedures (20). For symptomatic 7

pyelonephritis fluconazole is recommended (14 days) or amphotericin B deoxycholate if 8

resistance to fluconazole is suspected (1 to 7 days)(20). 9

10

In the setting of oesophageal candidiasis systemic therapy is always recommended in HIV 11

patients (56), based on current published guidelines (20). A diagnostic trial of antifungal 12

therapy (14 to 21 days) is reasonable before performing endoscopy (57) and oral or 13

intravenous fluconazole, or an echinocandin are the treatment options. In ICU patients, 14

there is no evidence at the present to treat with systemic antifungal therapy. 15

16

Candida endocarditis requires aggressive treatment and a high index of suspicion (58). High-17

dose echinocandin is recommended as initial therapy, although the evidence for this is weak 18

(20,59): caspofungin 150mg daily, micafungin 150mg daily, or anidulafungin 200 mg daily. 19

Liposomal amphotericin B, 3-5 mg/kg daily, with or without flucytosine, 25 mg/kg 4 times 20

daily, is an alternative(60). Step-down therapy to fluconazole, 400-800 mg (6-12 mg/kg) 21

daily is recommended when susceptible isolates and negative blood cultures are obtained. 22

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

15

Whenever possible, valve replacement should be performed and treatment continued for at 1

least six weeks after surgery. 2

3

In Candida meningitis a combination of liposomal amphotericin B, 5 mg/kg daily, with oral 4

flucytosine, 25 mg/kg 4 times daily, is recommended, although evidence for this is weak(20). 5

Amphotericin B can also be used alone. Fluconazole, 400-800 mg (6-12 mg/kg) daily is 6

recommended for step-down therapy. This can be considered after the patient has 7

responded to initial therapy. The duration of treatment is not known: therapy is continued 8

until all signs, symptoms, cerebro-spinal fluid and radiological abnormalities have resolved. If 9

a central nervous system device is in place it should be removed if possible. 10

11

In figure 1 an algorithm is presented based on clinical evaluation, Candida scores and 12

biomarkers. Measurement of drug concentrations in the plasma should be performed when 13

treating with voriconazole and flucytosine. This is not routinely recommended for other 14

agents (61). This is particularly important in patients requiring renal replacement therapy. 15

16

17

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

16

CONCLUSION 1

The mortality of invasive candidiasis can be as low as 5%, but may be as high as 70% (9,10). 2

Early diagnosis and development of newer diagnostic technologies with higher sensitivity 3

and specificity is of crucial importance. Prophylactic therapy needs to be better evaluated 4

and is supported only by weak evidence. Empirical treatment is still a matter of controversy 5

but generally involves the use of an echinocandin and, when possible, a step-down therapy 6

with fluconazole. Source control is an important part of treatment - particularly in the case 7

of candidemia related to intravascular catheters. Diagnostic parameters must be 8

consistently adhered to in order to guide treatment and future research. 9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

17

REFERENCES 1 2 1. Valerio M, Rodriguez-gonzalez CG, Mun P, Caliz B, Sanjurjo M, Bouza E. Evaluation of 3

antifungal use in a tertiary care institution : antifungal stewardship urgently needed. J 4 Antimicrob Chemother. 2014;69(March):1993–9. 5

2. Nivoix Y, Launoy A, Lutun P, Moulin J, Pang KP, Leve D, et al. Adherence to 6 recommendations for the use of antifungal agents in a tertiary care hospital. J 7 Antimicrob Chemother. 2012;67(July):2506–13. 8

3. Montravers P, Dupont H, Gauzit R, Veber B, Auboyer C, Blin P, et al. Candida as a risk 9 factor for mortality in peritonitis. Crit Care Med. United States; 2006 Mar;34(3):646–10 52. 11

4. Montravers P, Mira J-P, Gangneux J-P, Leroy O, Lortholary O. A multicentre study of 12 antifungal strategies and outcome of Candida spp. peritonitis in intensive-care units. 13 Clin Microbiol Infect. England; 2011 Jul;17(7):1061–7. 14

5. Kollef M, Micek S, Hampton N, Doherty JA, Kumar A. Septic shock attributed to 15 Candida infection: importance of empiric therapy and source control. Clin Infect Dis. 16 United States; 2012 Jun;54(12):1739–46. 17

6. Loeches IM, Perner A. Focus on infection and sepsis in intensive care patients. 18 Intensive Care Med. Springer Berlin Heidelberg; 2016;42(4):491–3. 19

7. Magill SS, Edwards JR, Bamberg W, Beldavs ZG, Dumyati G, Kainer M, et al. Multistate 20 Point-Prevalence Survey of Health Care–Associated Infections. N Engl J Med. 21 2014;370(13):1198–208. 22

8. Pfaller MA, Andes DR, Diekema DJ, Horn DL, Reboli AC, Rotstein C, et al. Epidemiology 23 and Outcomes of Invasive Candidiasis Due to Non-albicans Species of Candida in 2 , 24 496 Patients : Data from the Prospective Antifungal Therapy ( PATH ) Registry 2004 – 25 2008. PLoS One. 2014;9(7):e101510. 26

9. Bassetti M, Ansaldi F, Diaz-martin A, Trecarichi EM, Garnacho-montero J, Rocca G 27 Della, et al. A multicenter study of septic shock due to candidemia : outcomes and 28 predictors of mortality. Intensive Care Med. 2014;40:839–45. 29

10. Horn DL, Ostrosky-Zeichner L, Morris MI, Ullmann AJ, Wu C, Buell DN, et al. Factors 30 related to survival and treatment success in invasive candidiasis or candidemia: a 31 pooled analysis of two large, prospective, micafungin trials. Eur J Clin Microbiol {&} 32 Infect Dis [Internet]. 2010 Feb;29(2):223–9. 33

11. Underhill DM, Iliev ID. The mycobiota: interactions between commensal fungi and the 34 host immune system. Nat Rev Immunol [Internet]. Nature Publishing Group, a division 35 of Macmillan Publishers Limited. All Rights Reserved.; 2014 Jun;14(6):405–16. 36

12. Madach K, Kristof K, Tualassay E, Ivanyi Z, Erdei A, Kiraly A, et al. Mucosal Immunity 37 and the Intestinal Microbiome in the Development of Critical Illness. Int Sch Res Netw. 38 2011;2011(545729):12. 39

13. Blumberg HM, Jarvis WR, Soucie JM, Edwards JE, Patterson JE, Pfaller MA, et al. Risk 40 factors for candidal bloodstream infections in surgical intensive care unit patients: 41 the NEMIS prospective multicenter study. The National Epidemiology of Mycosis 42 Survey. Clin Infect Dis. United States; 2001 Jul;33(2):177–86. 43

14. Mayer FL, Wilson D, Hube B, Mayer FL, Wilson D, Hube B, et al. Candida albicans 44 pathogenicity mechanisms Candida albicans pathogenicity mechanisms. Virulence. 45 2013;4(2):119–28. 46

15. Lafleur MD, Kumamoto CA, Lewis K. Candida albicans Biofilms Produce Antifungal-47

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

18

Tolerant Persister Cells. Antimicrob Agents Chemother. 2006;50(11):3839–46. 1 16. Lewis RE, Cahyame-zuniga L, Leventakos K, Chamilos G, Ben-ami R, Tamboli P, et al. 2

Epidemiology and sites of involvement of invasive fungal infections in patients with 3 haematological malignancies: a 20-year autopsy study. Mycoses. 2013;56:638–45. 4

17. Rammaert B, Desjardins A, Lortholary O. New insights into hepatosplenic candidosis, a 5 manifestation of chronic disseminated candidosis. Mycoses [Internet]. Blackwell 6 Publishing Ltd; 2012;55(3):e74--e84. 7

18. Gow NAR, van de Veerdonk FL, Brown AJP, Netea MG. Candida albicans 8 morphogenesis and host defence: discriminating invasion from colonization. Nat Rev 9 Micro [Internet]. Nature Publishing Group, a division of Macmillan Publishers Limited. 10 All Rights Reserved.; 2012 Feb;10(2):112–22. 11

19. Oude Lashof AML, Rothova A, Sobel JD, Ruhnke M, Pappas PG, Viscoli C, et al. Ocular 12 Manifestations of Candidemia. Clin Infect Dis [Internet]. 2011;53(3):262. 13

20. Pappas PG, Kauffman CA, Andes DR, Clancy CJ, Marr KA, Ostrosky-Zeichner L, et al. 14 Clinical Practice Guideline for the Management of Candidiasis: 2016 Update by the 15 Infectious Diseases Society of America. Clin Infect Dis [Internet]. 2016;62(4):e1. 16

21. Gamaletsou MN, Kontoyiannis DP, Sipsas N V, Moriyama B, Alexander E, Roilides E, et 17 al. Candida Osteomyelitis : Analysis of 207 Pediatric and Adult Cases ( 1970 – 2011 ). 18 Clin Infect Dis. 2012;55(10):1338–51. 19

22. Arnold CJ, Johnson M, Bayer AS, Bradley S, Giannitsioti E, Miró JM, et al. Candida 20 Infective Endocarditis : an Observational Cohort Study with a Focus on Therapy. 21 2015;59(4):2365–73. 22

23. O’Brien D, Stevens NT, Lim CH, O’Brien DF, Smyth E, Fitzpatrick F, et al. Candida 23 infection of the central nervous system following neurosurgery: a 12-year review. 24 Acta Neurochir (Wien) [Internet]. 2011 Jun;153(6):1347–50. 25

24. Pfaller MA, Rex JH, Rinaldi MG. Antifungal Susceptibility Testing: Technical Advances 26 and Potential Clinical Applications. Clin Infect Dis [Internet]. 1997;24(5):776. 27

25. WHO. Antimicrobial resistance - Global report on Surveillance. 2014. 28 26. Martín-Mazuelos E, Loza A, Castro C, Macías D, Zakariya I, Saavedra P, et al. β-D-29

Glucan and Candida albicans germ tube antibody in ICU patients with invasive 30 candidiasis. Intensive Care Med. 2015 Aug;41(8):1424–32. 31

27. Chumpitazi BFF, Lebeau B, Faure-Cognet O, Hamidfar-Roy R, Timsit J-F, Pavese P, et al. 32 Characteristic and clinical relevance of Candida mannan test in the diagnosis of 33 probable invasive candidiasis. Med Mycol [Internet]. 2014;52(5):462. 34

28. Kourkoumpetis TK, Fuchs BB, Coleman JJ, Desalermos A, Mylonakis E. Polymerase 35 Chain Reaction–Based Assays for the Diagnosis of Invasive Fungal Infections. Clin 36 Infect Dis [Internet]. 2012;54(9):1322. 37

29. Lacroix C, Gicquel A, Sendid B, Meyer J, Accoceberry I, Franc N. Evaluation of two 38 matrix-assisted laser desorption ionization-time of flight mass spectrometry ( MALDI-39 TOF MS ) systems for the identification of Candida species. Clin Microbiol Infect. 40 2014;20(2):153–8. 41

30. Hall L, Febre KM Le, Deml SM, Wohlfiel SL, Wengenack NL, Yeast T, et al. Evaluation of 42 the Yeast Traffic Light PNA FISH Probes for Identification of Candida Species from 43 Positive Blood Cultures. J Clin Microbiol. 2012;50(4):1446–8. 44

31. De Pauw B, Walsh TJ, Donnelly JP, Stevens DA, Edwards JE, Calandra T, et al. Revised 45 Definitions of Invasive Fungal Disease from the European Organization for Research 46 and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the 47

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

19

National Institute of Allergy and Infectious Diseases Mycoses Study Group 1 (EORTC/MSG) C. Clin Infect Dis [Internet]. 2008;46(12):1813. 2

32. Kullberg BJ, Arendrup MC. Invasive Candidiasis. N Engl J Med [Internet]. 3 2015;373(15):1445–56. 4

33. Pfaller MA, Diekema DJ. Epidemiology of Invasive Candidiasis : a Persistent Public 5 Health Problem. Clin Microbiol Rev. 2007;20(1):133–63. 6

34. Vallabhaneni S, Mody RK, Walker T, Chiller T. The Global Burden of Fungal Diseases. 7 Infect Dis Clin [Internet]. Elsevier; 2017 Jul 10;30(1):1–11. 8

35. Vincent J, Rello J, Marshall J. International study of the prevalence and outcomes of 9 infection in intensive care units. JAMA [Internet]. 2009;302(21):2323–9. 10

36. Garey KW, Rege M, Pai MP, Mingo DE, Suda KJ, Turpin RS, et al. Time to Initiation of 11 Fluconazole Therapy Impacts Mortality in Patients with Candidemia: A Multi-12 Institutional Study. Clin Infect Dis [Internet]. 2006;43(1):25. 13

37. Pappas PG, Kauffman CA, Andes D, Benjamin JDK, Calandra TF, Edwards JJE, et al. 14 Clinical Practice Guidelines for the Management Candidiasis: 2009 Update by the 15 Infectious Diseases Society of America. Clin Infect Dis [Internet]. 2009;48(5):503. 16

38. Eggimann P, Pittet D. Candida colonization index and subsequent infection in critically 17 ill surgical patients : 20 years later. Intensive Care Med. 2014;40:1429–48. 18

39. Leon C, Ruiz-Santana S, Saavedra P, Almirante B, Nolla-Salas J, Alvarez-Lerma F, et al. 19 A bedside scoring system (“Candida score”) for early antifungal treatment in 20 nonneutropenic critically ill patients with Candida colonization. Crit Care Med. 21 2006;34(3):730–7. 22

40. León C, Ruiz-Santana S, Saavedra P, Galván B, Blanco A, Castro C, et al. Usefulness of 23 the “Candida score” for discriminating between Candida colonization and invasive 24 candidiasis in non-neutropenic critically ill patients: A prospective multicenter study. 25 Crit Care Med [Internet]. 2009;37(5). 26

41. Leroy G, Lambiotte F, Thévenin D, Lemaire C, Parmentier E, Devos P, et al. Evaluation 27 of “Candida score” in critically ill patients: a prospective, multicenter, observational, 28 cohort study. Ann Intensive Care [Internet]. 2011;1(1):50. 29

42. Azoulay E, Dupont H, Tabah A, Lortholary O, Stahl J-P, Francais A, et al. Systemic 30 antifungal therapy in critically ill patients without invasive fungal infection*. Crit Care 31 Med. United States; 2012 Mar;40(3):813–22. 32

43. Cornely OA, Bassetti M, Calandra T, Garbino J, Kullberg BJ, Lortholary O, et al. 33 ESCMID* guideline for the diagnosis and management of Candida diseases 2012: non-34 neutropenic adult patients. Clin Microbiol Infect. England; 2012 Dec;18 Suppl 7:19–37. 35

44. Playford E, Lipman J, Sorrell T. Prophylaxis, empirical and preemptive treatment of 36 invasive candidiasis. CurrOpinCrit Care. 2010;16(5):470–4. 37

45. Cortegiani A, Russotto V, Maggiore A, Attanasio M, Naro AR, Raineri SM, et al. 38 Antifungal agents for preventing fungal infections in non-neutropenic critically ill 39 patients. Cochrane database Syst Rev. England; 2016 Jan;(1):CD004920. 40

46. Bailly S, Bouadma L, Azoulay E, Orgeas MG, Adrie C, Souweine B, et al. Failure of 41 Empirical Systemic Antifungal Therapy in Mechanically Ventilated Critically Ill Patients. 42 Am J Respir Crit Care Med [Internet]. 2015;191(10):1139–46. 43

47. Bassetti M, Bouza E. Invasive mould infections in the ICU setting: Complexities and 44 solutions. J Antimicrob Chemother. 2017;72(October):i39–47. 45

48. Mikulska M, Calandra T, Sanguinetti M, Poulain D, Viscoli C. The use of mannan 46 antigen and anti-mannan antibodies in the diagnosis of invasive candidiasis: 47

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

20

recommendations from the Third European Conference on Infections in Leukemia. 1 Crit Care. 2010;14(6):R222. 2

49. De Pascale G, Tumbarello M. Fungal infections in the ICU: advances in treatment and 3 diagnosis. Curr Opin Crit Care. United States; 2015 Oct;21(5):421–9. 4

50. Eggimann P, Garbino J, Pittet D. Management of Candida species infections in critically 5 ill patients. Lancet Infect Dis. 2003;3(12):772–85. 6

51. Piarroux R, Grenouillet F, Balvay P, Tran V, Blasco G, Millon L, et al. Assessment of 7 preemptive treatment to prevent severe candidiasis in critically ill surgical patients(*). 8 Crit Care Med [Internet]. 2004;32(12):2443–9. 9

52. Ostrosky-Zeichner L, Alexander BD, Kett DH, Vazquez J, Pappas PG, Saeki F, et al. 10 Multicenter Clinical Evaluation of the (1->3) -D-Glucan Assay as an Aid to Diagnosis 11 of Fungal Infections in Humans. Clin Infect Dis [Internet]. 2005;41(5):654–9. 12

53. Hanson KE, Pfeiffer CD, Lease ED, Balch AH, Zaas AK, Perfect JR, et al. Β-D-glucan 13 surveillance with preemptive anidulafungin for invasive candidiasis in intensive care 14 unit patients: A randomized pilot study. PLoS One. 2012;7(8):1–8. 15

54. Dupont H, Bourichon A, Paugam-Burtz C, Mantz J, Desmonts J-M. Can yeast isolation 16 in peritoneal fluid be predicted in intensive care unit patients with peritonitis? Crit 17 Care Med. United States; 2003 Mar;31(3):752–7. 18

55. Lindau S, Nadermann M, Ackermann H, Bingold TM, Stephan C, Kempf VAJ, et al. 19 Antifungal therapy in patients with pulmonary Candida spp. colonization may have no 20 beneficial effects. J intensive care. England; 2015;3(1):31. 21

56. Pace F, Antinori S, Repici A. What is new in esophageal injury (infection, drug-induced, 22 caustic, stricture, perforation)? Curr Opin Gastroenterol. 2009;25(4):372–9. 23

57. Wilcox CM, Alexander LN, Clark WS, Thompson SE 3rd. Fluconazole compared with 24 endoscopy for human immunodeficiency virus-infected patients with esophageal 25 symptoms. Gastroenterology. United States; 1996 Jun;110(6):1803–9. 26

58. Pasha AK, Lee JZ, Low SW, Desai H, Lee KS, Al Mohajer M. Fungal Endocarditis: Update 27 on Diagnosis and Management. Am J Med [Internet]. Elsevier Inc; 2016;129(10):1037–28 43. 29

59. Talarmin JP, Boutoille D, Tattevin P, Abgueguen P, Ansart S, Roblot F, et al. Candida 30 endocarditis: Role of new antifungal agents. Mycoses. 2009;52(1):60–6. 31

60. Baddley JW, Benjamin DK, Patel M, Miró J, Athan E, Barsic B, et al. Candida infective 32 endocarditis. Eur J Clin Microbiol Infect Dis. 2008;27(7):519–29. 33

61. Ashbee HR, Barnes RA, Johnson EM, Richardson MD, Gorton R, Hope WW. 34 Therapeutic drug monitoring (TDM) of antifungal agents: Guidelines from the british 35 society for medical mycology. J Antimicrob Chemother. 2014;69(5):1162–76. 36

37 38

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 1. Main risk factors of invasive candidiasis

High APACHE II score

Diabetes mellitus

Neutropenia

Renal insufficiency

Surgery, mainly abdominal

Pancreatitis

Use of broad-spectrum antibiotics

Parenteral nutrition

Hemodialysis

Mechanical ventilation

Presence of central venous catheters

Immunosuppressive agents

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 2. The most relevant diagnostic tests for invasive Candida infections (IC)

BDG: (1,3)-β-D-glucan assay; CAGTA: Candida albicans germ tube antibody; IVIG:

intravenous immunoglobulin

Diagnostic

test

Advantages Limitations

Culture • Gold standard

• Required for antifungal

susceptibility testing

• Results take time

• Some infecting organisms do

not grow on cultures

• Insensitive, may identify only ~

50% of all patients with IC (61)

Histology • Gold standard

• Broad categories of

potential pathogens

can be differentiated

• Invasive procedure

• Sensitivity depends on proper

sample collection

• It has limited specificity to

identify genus and species level

Antigen

detection:

(1,3)-β-D-

glucan

assay

(serum)

• Serial surveillance of

at-risk patients may

help early

identification of IC and

start preemptive

therapy (except for

hematologic

• Nonspecific, panfungal

screening test. BDG is a cell-

wall constituent of Candida

spp, Aspergillus spp,

Pneumocystis jirovecii and

several other fungi.

• Repeated positive results are

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

malignancy patients in

whom false positive

results are common

(62))

• May be used for

monitoring response to

therapy

required – the utility of a single

test result for diagnosis is very

limited

• False-positive tests are

common in the case of certain

haemodialysis membranes,

IVIG, surgical gauze products,

possible in presence of beta

lactam antibiotics (62)

• False-negative results can be

caused by haemolysis, high

cut-off values (80 vs. 350 vs.

800pg/ml) [29]

Antibody

detection:

CAGTA

assay

(serum)

• Helps to identify

previous exposures and

at-risk patients, may

indicate antifungal

prophylaxis prior to

immunosuppressive

therapy

• Not affected by

Candida colonization or

antifungal treatment

• Does not differentiate between

previous and active disease

• Sensitivity of an antibody test

is limited for

immunocompromised patients,

who often can not produce

antibodies against Candida spp

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

[30]

Nucleic

acid

detection

• Rapid results

• Sensitivity and

specificity can be high

with certain specimens

• Lack of standardized

methodologies

• It cannot differentiate between

infection, colonization or

contamination

• Contamination can be a

frequent problem

• Most assays require special

equipment and expertise

• Can be expensive

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 3. Treatment of candidemia

Treatment for Candidemia (Based on 2016 IDSA recommendations)

Non neutropaenic patient Neutropaenic patient

1) An echinocandin as initial

therapy

- Caspofungin: 70 mg then 50

mg daily

- Micafungin: 100 mg daily

- Anidulafungin: 200 mg then

100 mg daily

1) An echinocandin as initial

therapy (see non neutropaenic

patients).

2) Fluconazole: 12 mg/kg then 6

mg/kg daily, not in critically ill

and only when resistance is

unlikely. Recommended as step-

down therapy when

susceptibility is established (5-7

days). Negative repeated blood

cultures must be documented.

For C. glabrata use 12 mg/kg

dailya.

2) Lipid formulation amphotericin B

(3-5 mg/kg) is a less attractive

alternative.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

3) Lipid formulation amphotericin B

(3-5 mg/kg) is an alternative in

case of intolerance or resistance

to other agents.

3) Fluconazole (see non neutropenic

patients), not in critically ill and only

if no prior exposure. Can be used as

step-down therapy.

4) Voriconazole 6 mg/kg for 2

doses and then 3-4 mg/kg, can

be used as step-down therapy.

a For fluconazole-resistant C. glabrata, AmB deoxycholate, 0.3–0.6 mg/kg daily for 1–

7 days OR oral flucytosine, 25 mg/ kg 4 times daily for 7–10 days is recommended.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 4. Spectrum of activity of available systemic antifungal agents. Table incorporating the present ESCMID/IDSA (European Society of Clinical Microbiology and Infectious Diseases/ (Infectious Diseases Society of America) guidelines, modified after Nett et al. (reference: Nett JE, Andes DR. Antifungal Agents: Spectrum of Activity, Pharmacology, and Clinical Indications. Infect Dis Clin North Am. 2016 Mar;30(1):51-83. doi: 10.1016/j.idc.2015.10.012.)

5FC: flucytosine; AMB: amphotericin B; ANI: anidulafungin; CAS: caspofungin; FLU: fluconazole; ISA: isavuconazole; ITR: itraconazole; MICA: micafungin; POS: posaconazole; VOR: voriconazole; D: dimorphic fungus; M: mold; Y: yeast; green background: potent antifungal activity; orange background: increasing fungal resistance; red background definitive fungal resistance; CNS: central nervous system

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

Figure 1 1

ICU patients with signs of sepsis (CI : colonization index. CS : Candida score). A high rate 2

of invasive candidiasis is defined when there is >10% risk, as compared with the normal rates 3

of 1-2%. 4

5

Candida scores

CI ≥ 0.5

or CS ≥ 3

CI < 0.5

or CS < 3

6

7

8

9

10

11

12

13

14

Start treatment -Continue if positive mannan/antimannan -Stop if negative biomarker

Consider treatment -If started, stop if mannan/antimannan negative

Consider treatment if mannan/antimannan positive

Risk assessment

high or low

Risk assessment

high

Risk assessment

low

Follow up