EXHALED BREATH ANALYSIS IN CHRONIC PANCREATITIS

Abstract

Introduction: Human exhaled breath contains a wide range of molecules either present as gases or occurring in solubilized form in the humidity of the breath. Aim: To evaluate whether breath analysis is able to differentiate chronic pancreatitis (CP) patients from healthy subjects (HS). Subjects - Thirty-one consecutive patients (25 males, 6 females; mean±SD age: 58±11 years, range: 37-78) with proven CP were studied. The pancreatitis was of alcoholic origin in 22 patients, obstructive in 2 and of unknown origin in 7. The diagnosis of CP was made according to the Rome-Marseilles criteria; it was based on clinical history and was confirmed by ultrasonography and/or contrast-enhanced computed tomography, magnetic resonance imaging and endoscopic retrograde cholangiopancreatography and/or histology. On the basis of fecal elastase-1 (ScheBo Tech, Germany; reference value: greater than 200 μg/g), 11 CP patients had exocrine pancreatic insufficiency. No patients had pain at the time of the study. Thirty-one HS (9 males, 22 females; mean age 52±18 years, range: 25-78 years, P = 0.251 vs. chronic pancreatitis patients) recruited from the medical staff were used as controls. Methods: After an overnight fasting, a breath sample from all subjects was obtained in the same controlled clean room according to the manufacturer instruction. The subjects exhaled into a vial that was crimped afterwards. An ambient air sample was also obtained immediately after each patient examination. The exhaled patient samples, as well as the ambient air samples, were immediately stored at −20°C until analysis which was carried out using a mass spectrometer (Airsense, V&F Medical Development GmbH, Absam, Austria). Ninety-seven different substances were analyzed on the patients and ambient air samples. The data were expressed in ppb and were analyzed by means of the Mann-Whitney U-test. Two-tailed P values of less than 0.05 were considered statistically significant. Results: Considering the exhaled breath, 20 undefined substances with molecular masses 56, 84, 94, 95, 96, 97, 98, 99, 100, 101, 102, 104, 105, 106, 110, 112, 113, 114, 115, 116 were significantly lower in CP patients than in the controls, whereas the substance with molecular mass 66 was significantly higher in CP patients than in controls. Surprisingly, the ambient air collected was significantly different between CP patients and controls as regards the same substances detected in the exhaled breath of the patients, except for substances with molecular mass 66 and 102. Finally, no significant differences were found in CP patients between those with and those without pancreatic insufficiency. Conclusions: The results of this study show that CP patients exhaled some substances differently than HS. There is the need to identify these substances as well as to collect the exhaled breath in a closed circuit to prevent the possible biases due to ambient air. A provocative meal test coupled with this technique might be useful in order to evaluate exocrine pancreatic insufficiency in CP patients.