High level of β-(1,3)-D-glucan antigenaemia in cystic fibrosis in the absence of invasive fungal disease

Vilma Rautemaa1, Heather D Green2,3, Andrew M Jones2,3, Riina Rautemaa3,4

1 School of Medical Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Oxford Road, Manchester, M13 9PL, UK

2 Manchester Adult Cystic Fibrosis Centre, University Hospital of South Manchester, Southmoor Road, Manchester, M23 9LT, UK

3 Division of Infection, Immunity and Respiratory Medicine, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, The University of Manchester, Southmoor Road, Manchester, M23 9LT, UK

4 Mycology Reference Centre Manchester, University Hospital of South Manchester, Manchester, Southmoor Road, Manchester, M23 9LT, UK

Corresponding author:Dr Riina Rautemaa-Richardson DDS, PhD, FRCPath, PGCertMedEdClinical Senior Lecturer in Infectious Diseases and Medical Education; and Consultant in Medical Mycology

The University of Manchester, Division of Infection, Immunity and Respiratory Medicine, Manchester Academic Health Science Centre, University Hospital of South Manchester, Wythenshawe Hospital, Manchester, UK

Education and Research CentreWythenshawe HospitalSouthmoor RoadManchester, UKM23 9LT+44 (0)161 291 5941 riina.richardson@manchester.ac.uk

Word count: 3370

1

Abstract

β-(1,3)-D-glucan (BDG) is used to rule out invasive fungal disease (IFD) but its usefulness in

cystic fibrosis (CF) has not been evaluated. We measured serum BDG in CF patients with no

clinical suspicion of IFD. Samples from 46 adult CF patients during a stable period and during

pulmonary exacerbation were tested. The association of BDG with clinical variables was

analysed. 303 non-CF patients with suspected IFD were used as comparators. Both samples

were negative in 52% of CF patients, whereas 67% of comparators had only negative results

(p=0.08). CF patients with pancreatic insufficiency and CF-related diabetes had fewer

negative results (p<0.05 for both). Negative results were more common in older CF patients

(p<0.05). Use of antibiotics, presence of fungi in sputum and CF liver disease did not impact

BDG levels. In conclusion, patients with CF experience significant BDG antigenaemia in the

absence of IFD.

Key words:CF, beta glucan, systemic fungal infection, candidemia, fungal sepsis, fungemia

2

1. Introduction

Serum β-(1,3)-D-glucan (BDG) testing is recommended by the European Society of Clinical

Microbiology and Infectious Diseases (ESCMID) for the exclusion of invasive fungal diseases

(IFDs) in septic patients who do not respond to antibiotics.1 This has been shown to be safe

and effective in antifungal stewardship.2,3 BDG is a polysaccharide of glucose found in the cell

walls of most fungi, including Aspergillus, Candida, and Pneumocystis species. Healthy

individuals have very low serum BDG levels (typically 0-40 pg/ml), but in patients with IFD

the concentration increases as fungal cells shed BDG into the bloodstream.4 IFDs, especially

candidaemia, are difficult to treat and are associated with high mortality.5 At the same time,

many antifungal drugs have considerable side effects and a high cost, and their overuse is a

risk for the development of resistance.6 Thus, there is a clear clinical need for quick, reliable

and sensitive methods for the detection of fungal infection that can guide antifungal

treatment.

The BDG test has high sensitivity and high negative predictive value in low prevalence

settings.7 It is useful in ruling out candidaemia, aspergillosis and pneumocystosis, but fungal

species in the genera Cryptococcus, Blastomyces, Lichtheimia, Mucor, and Rhizopus either

lack BDG completely or produce very small amounts and thus may not be detected with this

method.8,9 Some sources of false positives have also been identified. For example, surgical

gauzes have been shown to release BDG and cause false positive serum BDG results.10

Elevated levels have also been reported in lung transplant recipients, ICU patients, and HIV

patients in the absence of IFD.11-13 BDG has also been detected in some antimicrobial agents

but these are an unlikely cause of false positives due to becoming diluted when distributed

3

in the body.14 High bilirubin and triglyceride levels may inhibit the BDG assay and cause false-

negative results.7

Cystic fibrosis (CF) is a genetic condition caused by mutations in the cystic fibrosis

transmembrane regulator (CFTR) gene that leads to dehydration of fluids in the respiratory

and gastrointestinal tracts, causing mucus plugging, blockage of fluid secretion,

inflammation, tissue damage and fibrosis. There is growing evidence that fungal pathogens

can have a significant impact on the clinical outcome of CF. Chronic colonisation of the lungs

by Aspergillus and Candida species has been linked with more rapid decline in lung function

and more frequent respiratory exacerbations.15,16 Patients colonised by Aspergillus may also

develop allergic bronchopulmonary aspergillosis (ABPA), a hypersensitivity reaction to the

fungus that is associated with decreased lung function.17 Despite our better understanding

of fungal respiratory diseases we still know very little about the level of fungal antigenaemia

in CF. In addition, permanent intravenous access devices are commonly used in CF due to

repeated need for venepuncture and IV antibiotics. Infection of these devices by fungal

pathogens is a common cause for their removal.

In our clinical experience, serum BDG test in septic CF patients is often reported positive in

the absence of any other evidence of IFD (e.g. negative blood culture, no response to

antifungal treatment). To our knowledge, there have been no studies assessing the use of

the BDG test in CF patients. Thus, the aim of this study was to determine the level of BDG

antigenaemia in adult CF patients who have no clinical suspicion of IFD, both during a period

of clinical stability and during a pulmonary exacerbation. We also recorded various clinical

variables to assess their relationship with BDG levels in these patients.

4

2. Methods

2.1. Patients

Paired serum and sputum samples were obtained from the Manchester Allergy, Respiratory

and Thoracic Surgery Biobank (ManARTS). Samples collected from 46 CF patients at the

Manchester Adult Cystic Fibrosis Centre at University Hospital of South Manchester (UHSM)

NHS Trust between 2013-2015 were included in the study. The study was approved by the

ManARTS ethics committee (REC reference: 10/H1010/7). None of the patients had any

clinical suspicion of IFD at either time point and did not receive systemic antifungal therapy.

Two pairs of serum and sputum samples were obtained for each patient; one serum and

sputum pair taken when the patient was stable, and another pair taken during a respiratory

exacerbation (total 92 serum and 92 sputum samples). To further analyse the variability of

BDG levels over the 13-month study period, 3-4 additional serum samples (mixture of stable

and exacerbation) were analysed for three patients with low BDG values and three patients

with high BDG values.

Respiratory exacerbation was diagnosed when oral or intravenous antibiotic treatment was

prescribed and the patient displayed in the previous seven days any four of the following

symptoms: increased cough, increased dyspnoea, change in sputum, new or increased

haemoptysis, malaise, fever >38°C, anorexia, sinus pain, new physical findings in the chest,

fall in spirometry by >10%, new radiographic findings (modified Fuchs criteria).18 All patients

were over 18 years old and had no history of HIV infection or transplantation, and were able

to spontaneously provide sputum. At the time of sampling, the median age of participants

was 30.0 years (range 21.1-65.7), and 59% of the patients were male. The median time

5

between the stable and exacerbation samples was 115.5 days (range 22-377). A spectrum of

clinical variables was recorded, including paired sputum fungal culture, white cell count

(WCC) and C-reactive protein (CRP), CF genotype, nebulised antibiotic use, oral steroid use,

lung function tests and presence of pancreatic insufficiency (PI), CF related diabetes (CFRD),

and CF liver disease (CFLD). The number of days on intravenous (IV) antibiotics in the year

2014 was also recorded.

2.2. Comparator group

A database search was performed of all serum BDG results for adult patients at UHSM

reported by the Mycology Reference Centre Manchester between 2013-2015; any results

from CF patients were excluded. The total number of comparator patients was 303 with 433

total samples. The median age of comparators at the time of sampling was 59.8 years (range

18.2-90.0), and 63% of the patients were male. At UHSM the BDG test is used to rule out a

diagnosis of IFD in septic patients who are non-responsive to broad-spectrum antibiotics and

have risk factors for invasive fungal disease. The test is not used for routine screening or to

rule in IFD. Thus, all patients in the comparator group had suspected IFD.

2.3. BDG assay

(1,3)-β-D-Glucan concentration was measured using the Fungitell® assay (Associates of Cape

Cod, East Falmouth, MA, USA) according to manufacturer’s instructions. Sputum samples

were diluted 1:100 with glucan-free water, serum samples were not diluted. 5 μl of each

sample was pre-treated with 20 μl of the alkaline reagent (final concentration 0.125 M KOH

and 0.6 M KCl) at 37°C for 10 minutes. Five concentrations of glucan standard were prepared

(range 31.25-500 pg/ml) and added to the plate along with a negative control. After this, 100

6

μl of the Fungitell® reagent was added to each well and the plate was kinetically monitored

at 405 nm minus 490 nm in an incubating plate reader (37°C) for 40 minutes. The mean rate

of optical density change between 0-40 minutes was calculated and BDG concentration

determined by using the standard curve. BDG values <60 pg/ml were interpreted as negative

(IFD unlikely), 60-79 pg/ml as indeterminate (IFD possible), and ≥80 pg/ml as positive (IFD

possible), as per kit instructions. Patients with negative BDG results only were classified as

“negative”, and those with one or more positive or indeterminate results were classified as

“not negative”. In addition, a sample of one commonly used short-acting insulin,

NovoRapid® Insulin Aspartate (100 units/ml) (Novo Nordisk, Bagsværd, Denmark) was

analysed for BDG.

2.4. Statistical methods

Statistical analysis was performed using SPSS Statistics version 22 (IBM Corporation, Armonk,

NY) and Graphpad Prism version 6 (GraphPad Software Inc, San Diego, CA). Stable and

exacerbation samples were analysed separately. BDG values were not normally distributed

so non-parametric tests were used for analysis. Spearman rank correlation was used to

analyse the association between stable and exacerbation BDG levels, between serum and

sputum BDG levels, and between BDG levels and the other continuous variables. Wilcoxon

matched pairs test was used to analyse the difference in BDG levels between paired stable

and exacerbation samples. Mann-Whitney U-test was used to analyse the association of BDG

levels with various categorical variables. The patients were divided into two groups based on

their serum BDG result (“negative” and “not negative”), and the Mann-Whitney test was

used for analysing the differences in the continuous variables between these two groups.

Chi-squared test was used to analyse differences in categorical variables between the

7

“negative and “not negative” groups. Chi-squared test was also used to analyse the

differences in the proportions of “negative” or “not negative” results between the CF group

and the control group. The Kruskall Wallis test was used for comparing the BDG levels of the

three different genotype groups. Any comparison of values with a level of significance,

p<0.05, were reported as statistically significant.

3. Results

3.1. Serum BDG levels

In the CF group the median serum BDG concentration was 45.7 pg/ml (range 0-472.3 pg/ml).

Overall 60% of stable and 59% of exacerbation samples were negative. Both stable and

exacerbation serum samples were negative for 24 (52%) of the 46 CF patients, one of the

samples was negative for 6 (13%) patients, and for 16 (35%) patients neither of the two

samples was negative. There was no significant difference between stable and exacerbation

serum BDG levels (median 40.2 vs. 48.7 pg/ml respectively, p=0.544), and the levels

correlated closely within patients (r=0.856, p<0.0001). In the suspected IFD group, the

median serum BDG concentration was 34.3 pg/ml (range <31.3 - >500 pg/ml). Overall 65% of

comparator samples were negative. All serum samples were negative for 204 (67%) of the

303 suspected IFD patients and 99 (33%) had at least one positive or indeterminate result.

There was no significant difference in the proportion of negative and not negative results

between the CF and suspected IFD groups (χ2=3.1, p=0.08).

8

Analysis of the additional serum samples (3-4 per patient) obtained from six CF patients

showed that the BDG levels remained relatively stable over the 13-month study period,

independent of respiratory exacerbations (Figure 1). Those three patients who were positive

at the start of the study period remained positive for the follow up period, and the three

who were negative at the start stayed negative.

All samples from pancreatic sufficient patients (n=7) were negative for BDG, whereas only

54% of stable samples and 51% exacerbation samples from patients with PI (n=39) were

negative (χ2=5.31 and 5.81 respectively, p<0.05 in both; Table 1). The median serum BDG

was higher in samples from patients with PI than from patients who were pancreatic

sufficient (exacerbation median 60.0 vs 17.6 pg/ml and stable median 55.3 vs. 25.3 pg/ml).

This difference was significant in exacerbation samples (p<0.05). Patients with CFRD (n=21)

were found to have significantly higher serum BDG than non-diabetic patients (median 82.3

vs. 30.6 pg/ml respectively, p<0.05 in stable samples and median 73.6 and 31.3 pg/ml

respectively, p<0.05 in exacerbation samples; Table 1). Non-diabetic patients had a negative

BDG result significantly more often than diabetic patients in both stable samples (62% and

38% negative respectively, χ2=8.414, p<0.005) and exacerbation samples (72% and 43%

negative respectively, χ2=3.998, p<0.05). To exclude an interaction between exogenous

insulin preparation and raised BDG levels we measured the concentration of BDG in one

commonly used short-acting insulin solutions, and this was very low (4.1 pg/ml).

CF patients with negative serum BDG were significantly older than patients with not negative

stable serum BDG result (34.5 and 26.7 years respectively, p<0.05) (Figure 2A). This

difference was not significant in exacerbation samples (36.5 vs. 27.5 years respectively,

9

p=0.117). BDG levels did not significantly correlate with age (Figure 2B). There was no

significant difference in the median age between the negative and not negative groups in the

suspected IFD patients (59.1 and 57.9 pg/ml respectively, p=0.893). Gender had no

significant impact on the BDG levels in either group.

3.2. Antibiotics and serum BDG

The median total number of days the CF patients were on IV antibiotics during the 12-month

follow up period was 20 (range 0-85 days). These consisted of home IVs (median 0 days;

range 0-62) and hospital IVs (median 13 days; range 0-85). In general, patients with high BDG

levels had less often required prolonged courses of IV antibiotics (Figure 3a). However,

there was no significant correlation between the number of days on IV antibiotics (total,

home or hospital) and serum BDG levels. Nor were there any significant differences in the

number of days on IV antibiotics between patients with negative and not negative serum

BDG results (Figure 3b).

The use of nebulised colistin was associated with higher serum BDG levels (Table 1). Patients

who were using colistin (n=26) had a significantly higher median exacerbation serum BDG

than patients not using colistin (64.9 and 21.5 pg/ml respectively, p<0.01). However, the

difference in stable BDG levels was not significant (59.3 and 30.4 pg/ml respectively,

p=0.144). The use of other nebulised antibiotics (tobramycin (n=8), aztreonam (n=9) and

meropenem (n=6)) did not impact on serum BDG levels at either time point.

3.3. Sputum BDG and culture findings

10

Median sputum BDG levels were >450x higher than the serum levels (median 18,260; range

0-182,303 pg/ml), with all but two samples reported as positive. There was no significant

correlation between paired sputum and serum BDG levels (p=0.076), and no significant

difference between stable and exacerbation sputum BDG levels (median 14,254 vs. 19,527

respectively, p=0.837). Patients with CFRD had significantly higher stable sputum BDG

concentration compared to non-diabetic patients (median 38,754 and 6,892 pg/ml

respectively, p<0.01). However, this difference was not observed in exacerbation samples

(median 20,768 and 18,235 pg/ml respectively, p=0.488).

Exacerbation sputum samples from patients with positive paired fungal sputum culture

(n=13) had significantly higher median BDG level than patients with negative sputum fungal

cultures (87,387 and 15,086 pg/ml respectively, p<0.01). In the stable samples there was no

difference in BDG levels between patients with positive (n=9) and negative sputum fungal

cultures (median 11,648 and 16,802 pg/ml respectively, p=0.870). Overall, six sputum

samples in the CF group were found to be positive for Pneumocystis jirovecii by PCR (2 stable

and 4 exacerbation samples). The median BDG of these samples (18,260.49 pg/ml) was not

different from the median of the rest of the group (18,076.89 pg/ml). Sputum fungal findings

did not correlate with serum BDG levels in either stable or exacerbation samples (Table 1a

and b).

There was no association between baseline FEV1 and FVC and serum or sputum BDG levels

in either stable or exacerbation samples (Table 2). Diagnosis of CFLD (n=17), use of port-a-

cath (n=16) or enteral feeding (n=9), CF genotype, serum CRP levels or white cell counts did

not correlate with BDG levels either (Tables 1 and 2).

11

4. Discussion

The results of this study show that patients with CF experience significant fungal

antigenaemia in the absence of invasive fungal disease. The serum BDG levels in CF patients

were comparable to those of patients on ICU with suspected IFD. To our knowledge, this has

not been reported before. This finding is of clinical significance in settings where the BDG

test is used in the diagnostics of IFD, as CF patients are likely to have false positive results

frequently.

It is not clear what the source of serum BDG in the CF patients with no evidence of IFD is, but

the gastrointestinal (GI) tract is one option. The chronically inflamed gut epithelium in CF is

highly permeable, allowing gut contents to leak into the bloodstream. 19-21 The gut houses

many species of fungi that are continuously shedding BDG, which may be responsible for

sustained serum antigenaemia we observed.22 Additionally, fungal pathogens are not the

only source of BDG in the body. β-D-glucans are a family of structurally related

polysaccharides of D-glucose that are found in mushrooms, cereal grains and some

bacteria.23,24 All these may react with the BDG assay if they reach the blood stream through

the intestinal wall. Therefore, abnormally elevated circulating BDG levels may serve as a

surrogate maker of intestinal damage. Interestingly, there was no association between

sputum fungal microbiological findings and serum BDG, indicating that translocation from

the lungs is an unlikely source of BDG in CF patients. This is supported by the lack of

correlation between sputum and serum BDG levels, as also described by Su et al in other

respiratory samples.25

12

Many CF patients suffer from pancreatic insufficiency and diabetes. Both type 1 and type 2

diabetes have been associated with increased intestinal permeability in non-CF patients.26

Our results showed that all patients who were pancreatic sufficient had negative serum BDG

results, whereas only half of patients with PI had negative results. Furthermore, patients

with CFRD had significantly higher serum BDG levels than non-diabetic patients. The

increased GI damage in PI and CFRD may increase the permeability of the gut epithelium,

allowing more BDG in. We tested a commonly used fast-acting insulin solution and found

that it contained very low levels of BDG. Thus it does not seem likely that insulin injections

are the source of BDG in diabetic patients, however further testing of the various insulin

preparations would be required to confirm this.

Serum BDG levels are determined not only by the amount entering the blood stream but

also by its clearance from it. BDG can be taken up and degraded by macrophages in the

spleen and lymph nodes, but the majority of circulating BDG is eliminated by the liver with

some contribution by the kidneys.27,28 BDG should be cleared within hours after entering the

bloodstream and should not remain stable over months, as in our patients. We found no

impact of CFLD on serum BDG levels, which suggests that in our patient population problems

with clearance are not the reason for high BDG levels.

BDG is highly immunogenic and is known to activate macrophages and the complement

system.29 Therefore, having consistently high levels of circulating BDG may promote systemic

inflammation, which may contribute to the clinical outcome of CF patients. This can be

potentiated by concomitant leakage of other microbial and dietary antigens. In line with

this, we found that patients with positive or indeterminate serum BDG were significantly

13

younger than patients with negative BDG. This could reflect the impact of BDG antigenaemia

on long-term survival.

There is some evidence that exposure to BDG potentiates the inflammatory response to

bacterial endotoxin and could worsen the outcome from bacterial infections.30,31

Interestingly, there was no association between IV antibiotic use and serum BDG levels. In

fact, our results showed that those patients with the highest BDG levels had very few days of

IV antibiotics during the follow-up. The similarity of BDG levels between stable and

exacerbation samples indicates that BDG is not linked to respiratory exacerbations. Perhaps

the pro-inflammatory effect of BDG provides some protection against infection.32

There are some limitations to this study. Firstly, the sample size was limited, which could

affect the statistical significance of our results. Secondly, as the BDG test has not been

validated for use with sputum samples, there were no previous publications or hospital

results to compare them to. Several studies have evaluated the use of the BDG test in

bronchoalveolar lavage fluid (BALF) for the diagnosis of IFD, and based on a recent meta-

analysis testing BALF for BDG is equivocal.33 Additionally, handling the viscous sputum

samples is difficult and pipetting small amounts accurately was very challenging, causing

variation between replicates.

Further work is needed to fully understand the cause and consequences of the high BDG

levels in CF. It has been shown that patients with CF have high serum levels of intestinal fatty

acid binding protein (iFABP), a reliable and specific marker for gut mucosal injury.34,35 This

could be an interesting marker to correlate with serum BDG levels, and might explain where

14

the BDG is coming from. A long-term follow-up of these patients may also elucidate the

impact of persistently high serum BDG levels and explore if BDG can be used as a marker for

disease prognosis or severity.

In conclusion, patients with CF experience significant BDG antigenaemia in the absence of

IFD. This is of clinical significance when using the BDG test for the diagnosis of IFD, as CF

patients are likely to have a significant number of false positive results. The systemic

inflammation caused by the antigenaemia may impact on the prognosis.

Acknowledgements

We thank Dr Malcolm Finkelman, Associates of Cape Cod, for his help and advice. We thank

Dr Philip Foden, University of Manchester Department of Medical Statistics, for his

assistance with the statistical analysis. We also thank the staff at the Mycology Reference

Centre Manchester for their assistance in performing the assay. In addition we thank the

study participants for their contribution.

Funding

This work was funded by a combined grant from the UK Cystic Fibrosis Trust and the British

Lung Foundation, and Associates of Cape Cod provided Fungitell kits. This report is

independent research supported by the National Institute for Health Research Clinical

Research Facility at University Hospital of South Manchester NHS Foundation Trust. The

views expressed in this publication are those of the authors and not necessarily those of the

NHS, the National Institute for Health Research or the Department of Health. The authors

15

would like to acknowledge the Manchester Allergy, Respiratory and Thoracic Surgery

Biobank and the North West Lung Centre Charity for supporting this project.

Conflicts of interest

The authors have no conflicts of interest to declare.

Tables and Figures

Figure 1. Serum BDG levels of six patients over a period of 13 months. BDG levels remained relatively stable over the follow-up period, independent of respiratory exacerbations. Those three patients who were positive at the start of the study period remained positive for the follow up period, and the three who were negative at the start stayed negative. Dashed line indicates threshold for negative result. Asterix (*) indicates stable samples, other samples were taken during pulmonary exacerbations.

a) b)

16

Figure 2. Association between age and stable serum BDG in CF patients. a) Patients with negative serum BDG had a significantly higher median age than patients with not negative (indeterminate or positive) serum BDG result (34.5 and 26.7 years respectively, p<0.05) b) Older patients tended to have lower BDG whereas younger patients had a wide range of BDG results, however this correlation was not statistically significant (r=-0.238, p=0.111). Dashed line indicates threshold for negative result.

a) b)

Figure 3. a) Correlation between serum BDG concentration and total number of days on IV antibiotics (both hospital and home IVs) during the 12-month follow-up period. There was no significant correlation between the two variables (hospital: p=0.256, home: p=0.384, total: p=0.939). Dashed line indicates threshold for negative result b) There was no significant difference in the total number of days on IV antibiotics in the 12-month follow-up period between patients with negative and not negative serum BDG results.

Table 2. Statistical analysis of categorical variables in stable (a) and exacerbation (b) serum samples from CF patients. Patients with CFRD had significantly higher BDG than non-diabetic patients in both stable and exacerbation samples. PI and use of colistin were associated with significantly higher BDG values in exacerbation samples but not in stable samples. None of the other variables had significant associations with BDG. PI: pancreatic insufficiency, CFRD: cystic fibrosis related diabetes, CFLD: cystic fibrosis liver disease, PJ PCR: Pneumocystis jirovecii polymerase chain reaction, neb: nebulised.a.

Variable Yes (n) No (n) Median BDG (Yes)

Median BDG (No) P value

PI 39 7 55.3 25.3 0.145CFRD 21 25 82.3 30.6 0.012CFLD 17 29 35.4 41.5 0.776Enteral feeding 9 37 47.7 37.1 0.879Port-a-cath 16 30 84.16 36.6 0.42Oral steroids 6 40 38.8 43.3 0.896Colistin (neb) 26 20 59.3 30.4 0.144Tobramycin (neb) 8 38 57.4 40.2 0.75Aztreonam (neb) 9 37 21.0 47.7 0.131

17

Meropenem (neb) 6 40 43.3 44.6 0.622Female 19 27 53.8 48.5 0.384Paired fungal culture 9 37 75.9 37.1 0.227Paired sputum PJ PCR 2 44 100.1 38.0 0.258GenotypeDelF508 homozygous 26 49.7 0.615DelF508 heterozygous 16 27.9Other 4 63.3

b.Variable Yes (n) No (n) Median

BDG (Yes)Median BDG

(No) P value

PI 39 7 60.0 17.6 0.012CFRD 21 25 73.6 31.3 0.042

CFLD 17 29 62.4 48.5 0.554

Enteral feeding 9 37 32.3 48.9 0.89

Port-a-cath 16 30 70.1 46.9 0.327

Oral steroids 6 40 50.3 48.7 0.935

Colistin (neb) 26 20 64.9 21.5 0.008

Tobramycin (neb) 8 38 57.9 46.9 0.919

Aztreonam (neb) 9 37 25.4 59.0 0.072

Meropenem (neb) 6 40 19.5 51.2 0.282

Female 19 27 53.4 48.5 0.384

Paired fungal culture 13 33 48.5 48.9 0.852

Paired sputum PJP PCR 4 42 45.9 48.7 1

Genotype

DelF508 homozygous 26 59.5 0.548

DelF508 heterozygous 16 34.4

Other 4 46.6

Table 3. Statistical analysis of continuous variables in stable and exacerbation serum samples from CF patients. None of the variables showed significant correlation with serum BDG results. CRP: C-reactive protein, WCC: white cell count, FEV1: forced expiratory flow 1 second, FVC: forced vital capacity

Variable Median (range) Correlation with serum BDG (p value)

Stable Exacerbation Stable Exacerbation

Age (years) 30.2 (21.1-65.7) 29.9 (20.9-65.9) 0.093 0.127

18

Paired WCC 9.2 (4.4-17.8) 11.4 (6.0-21.5) 0.647 0.370

Paired CRP 7.0 (1.0-75.0) 14.0 (1.0-192.0) 0.714 0.564

Baseline FEV1 (% predicted) 39.2 (21.1-86.7) 0.840 0.952

Baseline FVC (% predicted) 61.6 (37.9-104.1) 0.871 0.898

19

References

1 Cuenca-Estrella M, Verweij PE, Arendrup MC, et al. ESCMID* guideline for the diagnosis and management of Candida diseases 2012: diagnostic procedures. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases 2012;18 Suppl 7:9-18 doi: 10.1111/1469-0691.12038

2 Nucci M, Nouér SA, Esteves P, Guimarães T, Breda G, de Miranda BG, Queiroz-Telles F, Colombo AL. Discontinuation of empirical antifungal therapy in ICU patients using 1,3-β-d-glucan. J Antimicrob Chemother. 2016 Sep;71(9):2628-33. doi: 10.1093/jac/dkw188. Epub 2016 Jun 10.

3 Posteraro B, Tumbarello M, De Pascale G, Liberto E, Vallecoccia MS, De Carolis E, Di Gravio V, Trecarichi EM, Sanguinetti M, Antonelli M. (1,3)-β-d-Glucan-based antifungal treatment in critically ill adults at high risk of candidaemia: an observational study. J Antimicrob Chemother. 2016 Aug;71(8):2262-9. doi: 10.1093/jac/dkw112. Epub 2016 Apr 28.

4 Odabasi Z, Mattiuzzi G, Estey E, et al. Beta-D-glucan as a diagnostic adjunct for invasive fungal infections: validation, cutoff development, and performance in patients with acute myelogenous leukemia and myelodysplastic syndrome. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 2004;39(2):199-205 doi: 10.1086/421944

5 Zaoutis TE, Argon J, Chu J, et al. The epidemiology and attributable outcomes of candidemia in adults and children hospitalized in the United States: a propensity analysis. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 2005;41(9):1232-9 doi: 10.1086/496922

6 Peter T, Bissinger R, Signoretto E, et al. Micafungin-Induced Suicidal Erythrocyte Death. Cellular physiology and biochemistry : international journal of experimental cellular physiology, biochemistry, and pharmacology 2016;39(2):584-95 doi: 10.1159/000445650

7 Pickering JW, Sant HW, Bowles CA, et al. Evaluation of a (1->3)-beta-D-glucan assay for diagnosis of invasive fungal infections. Journal of clinical microbiology 2005;43(12):5957-62 doi: 10.1128/JCM.43.12.5957-5962.2005

8 Onishi A, Sugiyama D, Kogata Y, et al. Diagnostic accuracy of serum 1,3-beta-D-glucan for pneumocystis jiroveci pneumonia, invasive candidiasis, and invasive aspergillosis: systematic review and meta-analysis. Journal of clinical microbiology 2012;50(1):7-15 doi: 10.1128/JCM.05267-11

9 Girouard G, Lachance C, Pelletier R. Observations on (1-3)-beta-D-glucan detection as a diagnostic tool in endemic mycosis caused by Histoplasma or Blastomyces. Journal of medical microbiology 2007;56(Pt 7):1001-2 doi: 10.1099/jmm.0.47162-0

10 Nakao A, Yasui M, Kawagoe T, et al. False-positive endotoxemia derives from gauze glucan after hepatectomy for hepatocellular carcinoma with cirrhosis. Hepato-gastroenterology 1997;44(17):1413-8

11 Digby J, Kalbfleisch J, Glenn A, et al. Serum Glucan Levels Are Not Specific for Presence of Fungal Infections in Intensive Care Unit Patients. Clinical and Vaccine Immunology 2003;10(5):882-85 doi: 10.1128/cdli.10.5.882-885.2003

20

12 Alexander BD, Smith PB, Davis RD, et al. The (1,3){beta}-D-glucan test as an aid to early diagnosis of invasive fungal infections following lung transplantation. Journal of clinical microbiology 2010;48(11):4083-8 doi: 10.1128/JCM.01183-10

13 Morris A, Hillenbrand M, Finkelman M, et al. Serum (1-->3)-beta-D-glucan levels in HIV-infected individuals are associated with immunosuppression, inflammation, and cardiopulmonary function. Journal of acquired immune deficiency syndromes 2012;61(4):462-8 doi: 10.1097/QAI.0b013e318271799b

14 Marty FM, Lowry CM, Lempitski SJ, et al. Reactivity of (1-->3)-beta-d-glucan assay with commonly used intravenous antimicrobials. Antimicrobial agents and chemotherapy 2006;50(10):3450-3 doi: 10.1128/AAC.00658-06

15 Chotirmall SH, O'Donoghue E, Bennett K, et al. Sputum Candida albicans presages FEV(1) decline and hospital-treated exacerbations in cystic fibrosis. Chest 2010;138(5):1186-95 doi: 10.1378/chest.09-2996

16 Saunders RV, Modha DE, Claydon A, et al. Chronic Aspergillus fumigatus colonization of the pediatric cystic fibrosis airway is common and may be associated with a more rapid decline in lung function. Medical mycology 2016;54(5):537-43 doi: 10.1093/mmy/myv119

17 Fillaux J, Bremont F, Murris M, et al. Assessment of Aspergillus sensitization or persistent carriage as a factor in lung function impairment in cystic fibrosis patients. Scandinavian journal of infectious diseases 2012;44(11):842-7 doi: 10.3109/00365548.2012.695454

18 Fuchs HJ, Borowitz DS, Christiansen DH, et al. Effect of aerosolized recombinant human DNase on exacerbations of respiratory symptoms and on pulmonary function in patients with cystic fibrosis. The Pulmozyme Study Group. The New England journal of medicine 1994;331(10):637-42 doi: 10.1056/NEJM199409083311003

19 De Lisle RC, Borowitz D. The cystic fibrosis intestine. Cold Spring Harbor perspectives in medicine 2013;3(9):a009753 doi: 10.1101/cshperspect.a009753

20 Werlin SL, Benuri-Silbiger I, Kerem E, et al. Evidence of intestinal inflammation in patients with cystic fibrosis. Journal of pediatric gastroenterology and nutrition 2010;51(3):304-8 doi: 10.1097/MPG.0b013e3181d1b013

21 van Elburg RM, Uil JJ, van Aalderen WM, et al. Intestinal permeability in exocrine pancreatic insufficiency due to cystic fibrosis or chronic pancreatitis. Pediatric research 1996;39(6):985-91 doi: 10.1203/00006450-199606000-00010

22 Ellis M. Preventing microbial translocation in haematological malignancy. British journal of haematology 2004;125(3):282-93 doi: 10.1111/j.1365-2141.2004.04903

23 Zekovic DB, Kwiatkowski S, Vrvic MM, et al. Natural and modified (1-->3)-beta-D-glucans in health promotion and disease alleviation. Critical reviews in biotechnology 2005;25(4):205-30 doi: 10.1080/07388550500376166

24 McIntosh M, Stone BA, Stanisich VA. Curdlan and other bacterial (1-->3)-beta-D-glucans. Applied microbiology and biotechnology 2005;68(2):163-73 doi: 10.1007/s00253-005-1959-5

25 Su KC, Chou KT, Hsiao YH, Tseng CM, Su VY, Lee YC, Perng DW, Kou YR. Measuring (1,3)-β-D-glucan in tracheal aspirate, bronchoalveolar lavage fluid, and serum for detection of suspected Candida pneumonia in immunocompromised and critically ill patients: a prospective observational study. BMC Infect Dis. 2017 Apr 8;17(1):252. doi: 10.1186/s12879-017-2364-2.

21

26 de Kort S, Keszthelyi D, Masclee AA. Leaky gut and diabetes mellitus: what is the link? Obesity reviews : an official journal of the International Association for the Study of Obesity 2011;12(6):449-58 doi: 10.1111/j.1467-789X.2010.00845

27 Hong F, Yan J, Baran JT, et al. Mechanism by which orally administered beta-1,3-glucans enhance the tumoricidal activity of antitumor monoclonal antibodies in murine tumor models. Journal of immunology 2004;173(2):797-806

28 Yoshida M, Roth RI, Grunfeld C, et al. Soluble (1-->3)-beta-D-glucan purified from Candida albicans: biologic effects and distribution in blood and organs in rabbits. The Journal of laboratory and clinical medicine 1996;128(1):103-14

29 Kataoka K, Muta T, Yamazaki S, et al. Activation of macrophages by linear (13)-beta-D-glucans. Impliations for the recognition of fungi by innate immunity. The Journal of biological chemistry 2002;277(39):36825-31 doi: 10.1074/jbc.M206756200

30 Cook JA, Dougherty WJ, Holt TM. Enhanced sensitivity to endotoxin induced by the RE stimulant, glucan. Circ Shock 1980;7(3):225-38

31 Kikkert R, Bulder I, de Groot ER, et al. Potentiation of Toll-like receptor-induced cytokine production by (1-->3)-beta-D-glucans: implications for the monocyte activation test. Journal of endotoxin research 2007;13(3):140-9 doi: 10.1177/0968051907080024

32 Vetvicka V. Glucan-immunostimulant, adjuvant, potential drug. World journal of clinical oncology 2011;2(2):115-9 doi: 10.5306/wjco.v2.i2.115

33 Shi XY, Liu Y, Gu XM, et al. Diagnostic value of (1 --> 3)-beta-D-glucan in bronchoalveolar lavage fluid for invasive fungal disease: A meta-analysis. Respiratory medicine 2016; 117: 48-53 doi: 10.1016/j.rmed.2016.05.017

34 Timmermans K, Sir O, Kox M, et al. Circulating iFABP Levels as a marker of intestinal damage in trauma patients. Shock 2015;43(2):117-20 doi: 10.1097/SHK.0000000000000284

35 Adriaanse MP, van der Sande LJ, van den Neucker AM, et al. Evidence for a Cystic Fibrosis Enteropathy. PloS one 2015;10(10):e0138062 doi: 10.1371/journal.pone.0138062

22