Breath biomarkers of inflammation, infection and metabolic derangement in the intensive care unit

Sturney, Sharon C. (2015) Breath biomarkers of inflammation, infection and metabolic derangement in the intensive care unit. PhD thesis, University of Nottingham.

PDF (Thesis - as examined) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (2MB) | Preview


The analysis of volatile organic compounds (VOCs) in breath may be a useful non-invasive tool in the Intensive Care Unit (ICU) to monitor metabolic and oxidative stress or diagnose pulmonary infection. Acetone is produced during starvation and metabolic stress, hydrogen sulphide (H2S) may be a marker of inflammation and infection and hydrogen cyanide (HCN) may also act as a marker of infection, particularly caused by Pseudomonas aeruginosa.

Firstly, the effects on measured VOC concentrations of the breath collection equipment and storage were assessed. Sample humidity declined faster than any analyte. Sample losses of 21%, 25% and 24% for acetone, H2S and HCN, respectively, were seen as a result of being passed through the sampling apparatus. Over 90% of initial breath VOC concentrations were detectable after 90 min storage in Tedlar bags at 40°C.

Secondly, a breath collection method for off-line analysis was validated in 20 mechanically ventilated patients in the ICU. The effect on VOC concentrations of breath sampling from two locations after two breathing manoeuvres was explored, revealing significantly higher analyte concentrations in samples from the airways than from a T-piece in the breathing circuit, and after tidal breathing compared to a recruitment-style breath. Practical difficulties were encountered using direct airway sampling and delivering recruitment style breaths; end-tidal breath sampling from the T-piece was simplest to perform and results equally reproducible.

Breath samples from 26 healthy anaesthetised controls were used to validate a breath collection method in the operating theatre. The effects of altering anaesthesia machine settings on inspiratory and exhaled acetone concentrations were explored. A difference in median inspiratory, but not exhaled, acetone concentrations was observed between the anaesthesia machines (ADU Carestation 276 ppb, Aysis Carestation 131 ppb, p=0.0005). Closing the adjustable pressure limiting (APL) valve resulted in a reduction in exhaled acetone concentration, as did breath sampling distal to the circuit filter, due to dilution by dead space air. Median (range) breath concentrations for samples collected on the patient side of the circuit filter with the APL valve open (n=22): acetone 738 ppb (257–6594 ppb), H2S 1.00 ppb (0.71-2.49 ppb), HCN 0.82 ppb (0.60-1.51 ppb). Breath acetone concentration was related to plasma acetone (rs=0.80, p<0.0001) and beta-hydroxybutyrate concentrations (rs=0.55, p=0.0075).

Finally, breath and blood samples were collected daily from 32 mechanically ventilated patients in the ICU with stress hyperglycaemia (n=11) and/or new pulmonary infiltrates on chest radiograph (n=28). Samples were collected over a median 3 days (1-8 days). Median (range) breath VOC concentrations of all samples collected: acetone 853 ppb (162–11,375 ppb), H2S 0.96 ppb (0.22-5.12 ppb), HCN 0.76 ppb (0.31-11.5 ppb). Median initial breath acetone concentration was higher than in anaesthetised controls (1451 ppb versus 812 ppb; p=0.038). There was a trend towards a reduction in breath acetone concentration over time. Relationships were seen between breath acetone and arterial acetone (rs=0.64, p<0.0001) and beta-hydroxybutyrate (rs=0.52, p<0.0001) concentrations. Several patients remained ketotic despite insulin therapy and normal, or near normal, arterial glucose concentrations. Inspired and exhaled H2S and HCN concentrations were not significantly different. Breath H2S and HCN concentrations could not be used to differentiate between patients with pneumonia and those with pulmonary infiltrates due to other conditions.

In conclusion, losses due to the sampling apparatus were determined and linear over the range of concentrations tested. End-tidal breath sampling via the T-piece was the simplest technique, with reproducibility comparable to other methods. It was possible to replicate the breath sampling method in the operating theatre; pre-filter samples with inspiratory gas flow rate 6 L/min and APL valve open provided repeatable results avoiding rebreathing. There was no role for the use of breath H2S or HCN in the diagnosis or monitoring of pneumonia in critical illness. There was no relationship between breath acetone concentration and illness severity, however the utility of breath acetone in the modulation of insulin and feeding in critical illness merits further study.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Shaw, D.E.
Epton, M.
Keywords: Breath tests, Volatile organic compounds in breath, Breath sampling, Critical illness
Subjects: W Medicine and related subjects (NLM Classification) > WB Practice of medicine
Faculties/Schools: UK Campuses > Faculty of Medicine and Health Sciences > School of Medicine
Item ID: 28648
Depositing User: Sturney, Sharon
Date Deposited: 09 Oct 2015 10:35
Last Modified: 14 Oct 2017 12:33

Actions (Archive Staff Only)

Edit View Edit View