Community Research and Development Information Service - CORDIS


A collaborative trial was conducted to formally validate the effectiveness of an HPLC procedure for the determination of patulin in clear and cloudy apple juice and apple puree. The test portion of clear apple juice was directly extracted with ethyl acetate, whilst cloudy apple juice and apple puree were treated with enzymes prior to extraction. After back-extraction into sodium carbonate to remove interfering acidic compounds, the sample extract was dried, concentrated and patulin was determined by HPLC with UV detection. Clear and cloudy apple juice, and apple puree samples naturally contaminated with patulin and blank samples to which patulin was added were sent to 14 collaborators in 12 different European countries. Test portions of each of the three sample types were spiked with 75 ng/m l of patulin. Recoveries for patulin ranged from 80 to 92%. Based on the results for spiked samples (blind paired) as well as naturally contaminated samples (blind pairs at three levels) the relative standard deviation for repeatability (RSD) ranged from 8 to 35%. The relative standard deviation for reproducibility (RSD) ranged from 11 to 36%. Although HORRAT values of less than 1.4 were obtained for all three matrices at patulin levels ranging from 26 to 121 ng/mL, better performance values (RSD values ranging from 6 to 10% and RSD values ranging from 11 to25%) were obtained for clear and cloudy apple juice above 50 ng/mL which is the statutory advisory limit for patulin contamination in apple juices in many countries

Additional information

Bibliographic Reference: EUR 18857 EN (1999) 53pp.
Availability: Available from OOPEC Sales agents
ISBN: ISBN 92-828-5927-4
Record Number: 199910599 / Last updated on: 1999-04-01
Original language: en
Available languages: en