Evaluation of Faecal Volatile Organic Compounds in the Diagnosis of Paediatric Inflammatory Bowel Disease

Inflammatory bowel disease (IBD) is a growing and significant problem in childhood. The diagnosis and monitoring of IBD in children is highly invasive and is accompanied by risk, stress/anxiety for children and families and significant use of healthcare resources. Children, young people and families have told investigators that a non-invasive diagnostic test is a priority for the NHS.

IBD is an incurable chronic, relapsing condition of unknown aetiology. Its frequency has increased markedly across the world in recent years with the highest prevalence reported in Europe (322/100,000 population for CD; 505/100,000 for UC). A marked increased incidence has been seen in children especially in the Northern Hemisphere and largely due to an increase in Crohn's disease. Up to 25% of IBD starts in childhood or adolescence.

Early-onset IBD, possibly due to inheritance of a greater number of susceptibility genes and/or earlier exposure to environmental triggers, differs from that occurring in adults. In children, inflammation occurs most commonly in the colon complicating differentiation between CD and UC. Early-onset CD is more common in boys (whereas adult-onset disease occurs more commonly in females) and tends to be more extensive with greater involvement of the upper intestine. Colonic inflammation in children with UC also tends to be more extensive than in adults.

At initial presentation in all suspected children, upper and lower intestinal endoscopy (requiring two days of bowel cleansing and usually a general anaesthetic) and small bowel imaging by magnetic resonance enterography or wireless capsule endoscopy are required. Although serious adverse events are uncommon, these invasive investigations are highly stressful for children and their families and consume considerable hospital resources. Undertaking these investigations delays diagnosis and the start of treatment.

Accurate, non-invasive, rapid and cost-effective diagnostic tools that can be used in the clinic or ward setting and better means of differentiating CD from UC are needed. Faecal calprotectin (FC), a non-specific marker of intestinal inflammation, is recommended by NICE to distinguish IBD from non-inflammatory gut disorders such as irritable bowel syndrome (https://www.nice.org.uk/guidance/dg11) and a near-patient test has been developed (http://www.buhlmannlabs.ch/core/quantum-blue/calprotectin/). A recent meta-analysis (394 IBD/321 non-IBD controls) reported a pooled sensitivity and specificity of FC in the diagnosis of IBD in children of 0.98 (95% confidence interval: 0.95-1.0) and 0.68 (0.50- 0.86) respectively. The low specificity indicates that an alternative or additional test with higher specificity is needed to reduce the time to diagnosis and avoid many children undergoing unnecessary investigations.

There is great interest in the measurement of volatile organic compounds (VOCs) in breath and vapours from various human tissues (urine, stool, blood) in the diagnosis and monitoring of a wide range of diseases. VOCs emitted from stool are responsible for stool odour and consist of a large number of carbon based molecules of low molecular mass (<1.5Kd) including organic acids, alcohols, esters, heterocyclic compounds, aldehydes, ketones and alkanes. They result from the metabolism of the intestinal mucosa and the gut microbiota and their abundance changes according to the specific effects of intestinal diseases on these processes.

Investigators have demonstrated that faecal VOCs are relatively stable over time within and between individuals despite day-to-day variation in diet and are not affected by freezing and storage of stool samples. We have optimized procedures for stool handling for metabolite extraction for gas chromatography-mass spectrometry (GC-MS) analysis (sample volume, solid phase micro-extraction fibre coating, extraction conditions [temperature and time] and vial volume and also developed the advanced statistical methods required for determining and comparing VOC profiles.

Investigators have shown that the analysis of VOCs separates adult patients with diarrhoea-predominant irritable bowel syndrome (n=30) from those with active CD (62) and UC (n=48; sensitivity of 94% and 96% and specificity of 82% and 80% respectively; p<0.05). Our data in adults strongly suggests that VOCs differentiate CD from UC, and also Crohn's colitis from UC (Ahmed 2013). More recently, research has shown that VOCs also distinguish between active and inactive CD and UC in adult patients (Ahmed 2016).

Despite these encouraging findings, given the differences in IBD between adults and children and that gut microbiota changes with age, it cannot be assumed that diagnostic tests that are effective in adults will necessarily work in children. Therefore, this technology needs to be tested in children. In a proof of principle study, faecal VOCs differentiated children with IBD from healthy children both during active disease and remission. However, investigators are not aware of any previous studies in children that have evaluated the utility of faecal VOCs in differentiating IBD from other common gastrointestinal disorders in an out-patient setting or in monitoring response to treatment.

Measurement of VOCs in fresh stool samples using bench-top equipment in the clinic or ward will be possible in the near future. The University of Liverpool owns the intellectual property for the development of a point-of-care instrument. The development of this technology is supported by the University of Liverpool and Cancer Research UK (the latter for measurement of VOCs in urine to detect bladder and prostate cancer). Industrial partners are engaged in refining the prototype ready for moulding and production this year. The aim is to have a fully tested, CE marked, point-of-care instrument on sale in the EU for bladder and prostate cancer in 2020/1 and for IBD/irritable bowel syndrome in 2022.