The principal aim of this study is to establish if a polymorphism in a gene important for innate immunity to fungi represents a significant risk factor for the development of Candida colonisation and subsequent invasive candidosis in critically ill patients. Incorporation of a screening programme onto a risk-based algorithm for critical care patients would allow more effective targeting of molecular diagnostic tests, anti-fungal prophylaxis and targeted treatment. Sequential critical care patients will be screened for gene polymorphisms and undergo regular screening for Candida colonization.

Invasive fungal infections are associated with increased morbidity and mortality rates as high as 50% (1). Up to half occur in non-neutropenic; critically ill patients and the majority are due to Candida species. A number of randomised controlled trials (RCTs) have evaluated anti-fungal prophylaxis, predominantly with fluconazole in critically ill patients and three systematic reviews of these RCTs have been performed (2-4). Meta-analyses demonstrated a homogeneous effect of anti-fungal prophylaxis on the risk of proven infection and some suggested a reduction in mortality. However widespread use of anti-fungal drugs is expensive, promotes resistance, cause a shift in pathogenic species towards more difficult to treat infections and may be associated with adverse drug related events. It is necessary to establish a method to identify patients at greatest risk of fungal infection and thus those who will benefit most from anti-fungal prophylaxis. Numerous risk factors have been identified including colonization, class/duration of antibiotics, number/location/duration of lines, total parenteral nutrition, gastric acid suppression, bacterial sepsis, post-surgical complications etc, but none accurately predict infection or fungal related mortality. Given the risk factors for fungal colonisation include infection, sepsis and antibiotic use, information regarding concomitant antibody deficiency will also be collected using calculated globulin (part of routine liver function testing) and immunoglobulins.

Recent evidence suggests fungi can modulate and evade host immunity and use surface molecules coded for by immune response genes that enable them to promote protective tolerance leading to colonisation and persistence (5). Genetic predisposition to infection may be associated with polymorphisms in some of these immune response genes. Many of these polymorphisms relate to a single difference in only one single nucleotide at defined positions in the DNA sequence. These single nucleotide polymorphisms (SNPs) are common occurrences within the human genome although frequency can vary markedly according to ethnic or geographical background. A single SNP may alter both innate and adaptive immunity and increase the risk of invasive fungal disease.

Dectin-1 is a C-type lectin receptor that plays a central role in host defence against candidal infection. A common polymorphism in the gene coding for Dectin-1 has been identified and is associated a functional defect in the immune response and increased risk of candidal infection (6). Dectin-1 recognizes the ß-1,3-glucan on cell walls and synergizes with toll-like receptors (TLRs) TLR2 and TLR4 to promote Th1 and Th17 responses to activate anti-fungal host defence. It is suggested that Dectin-1 has a role in both resistance and tolerance to fungal infection (7). A polymorphism (Tyr238X) in exon 6 of the Dectin-1 gene, that results in an early stop codon leading to the loss of the last 10 amino acids of the extracellular domain, has been identified and leads to diminished binding capacity. This mutation is thought to be present in around 14% of the population with an allele frequency of 8% (8). Although data in UK populations are limited, heterozygosity for this mutation has been linked to recurrent mucocutaneous candidal infection and increased colonization rates (6). Although no link with invasive disease has been established yet, colonization is a major risk factor and so a causal link can be postulated. More studies are needed to investigate this and inform strategies for prophylaxis and pre-emptive anti-fungal treatment. Screening these patients for Dectin-1 polymorphism will not only investigate allele frequency in a UK population and provide information on this SNP as a risk factor for fungal infection but also enable this to be built into a rule-based algorithm that can identify high-risk patients who would benefit from anti-fungal prophylaxis and/or targeted molecular diagnostics.

The association between primary or acquired immunodeficiency and susceptibility to fungal infection has also long been recognised (9)and measurement of antibodies to Candida albicans have been used as a screening test for humoral immunodeficiency in patients (10). Perturbations of immunoglobulin levels have not been extensively studied in critically ill patients. However two small observational studies demonstrated that hypogammaglobulinaemia was relatively common in patients with septic shock on admission to critical care and was associated increased mortality (11 12). However, it is unclear whether this is a consequence of critical illness or a predisposition in some individuals that makes them more susceptible to critical illness. Calculated globulin (total protein - albumin) forms part of the liver function tests that are routinely performed in hospital and has been used on an all Wales basis to screen for previously undiagnosed primary and secondary immunodeficiency. Calculated globulin can be utilised in the same way to identify patients with low (<18g/dl) immunoglobulin levels at hospital or critical care admission. Changes in immunoglobulin can then be tracked throughout critical illness using the calculated globulin and additional analysis performed in selected patients for more subtle defects in humoral immunity.

Inclusion Criteria:

• Patients aged > 18 years
• Patients with anticipated or actual critical care unit stay > 72 hours
• Consent or assent (if patient lacks capacity) obtained

Exclusion Criteria:

• Age < 18 years
• Length of predicted stay on critical care < 72 hours consent or assent not obtained

Critical care patients predicted to stay more than 72 hours