Evaluation Report proficiency test. DLA 9/ Lactose and Fructose. "lactose free" food. Waldemar-Bonsels-Weg Ahrensburg, Germany

Transcription

1 DLA Dienstleistung Lebensmittel Analytik GbR Evaluation Report proficiency test DLA 9/2015 Lactose and Fructose in "lactose free" food Dienstleistung Lebensmittel Analytik GbR Waldemar-Bonsels-Weg Ahrensburg, Germany Coordinator of this PT: Dr. Matthias Besler Page 1 of 45

2 Inhalt / Content 1. Introduction Realisation Test material Homogeneity Test Submission of results Evaluation Assigned value Standard deviation Outliers Target standard deviation General model (Horwitz) Value by precision experiment Value by perception z-score z'-score Quotient Standard uncertainty Recovery rates: Spiking Results Fructose Fructose Sample A (in mg/100g) Fructose Sample B (in mg/100g) Fructose Spiking Material Sample (in g/100g) Lactose Qualitative Valuation Sample A and Sample B Lactose Sample B (in mg/100g) Lactose Spiking Material Sample (in g/100g) Recovery Rates for Lactose Galactose Galactose Sample A (in mg/100g) Galactose Sample B (in mg/100g) Galactose Spiking Material Sample (in g/100g) Documentation Primary data Homogeneity Analytical Methods Index of participant laboratories Index of references...45 Page 2 of 45

3 1. Introduction The participation in proficiency testing schemes is an essential element of the quality-management-system of every laboratory testing food and feed, cosmetics and food contact materials. The implementation of proficiency tests enables the participating laboratories to prove their own analytical competence under realistic conditions. At the same time they receive valuable data regarding the validity of the particular testing method. The purpose of DLA is to offer proficiency tests for selected parameters in concentrations with practical relevance. Realisation and evaluation of the present proficiency test follows the technical requirements of DIN EN ISO/IEC (2010) and DIN ISO 13528:2009 (6). 2. Realisation 2.1 Test material Two PT-samples for the detection of lactose/galactose and fructose and one spiking material sample were provided for analysis. The spiking material sample contains the respective analytes in the range of 1-10 % and was added to the spiked PT-sample. The results of the spiking material sample should give the possibility of a comparison with the spiked sample in respect to the detectability of the allergens with and without the influence of matrix and / or food processing. The test material consists of common in commerce "lactose free" cookies. These cookies contain among other ingredients "low in lactose" skimmed milk powder, which may contain hydrolyzed lactose (free galactose). The basic composition of both sample A and sample B was the same (see table 1). The spiking material, which contains lactose, fructose and galactose from a "lactose free" milk product, was added to sample B. The composition of the spiking material sample and the amounts of analytes in sample B are given in table 2. After homogenisation the samples were portioned to approximately 25 g. Page 3 of 45

4 Table 1: Composition of DLA-Sample Ingredients Sample A Sample B Cookies, lactose free Ingredients: Sugar, rice flour, banana flour, palm fat, potato starch, corn flour, glucose, corn flour gel, modified potato starch, inulin, low-lactose skimmed milk powder, salt, whole egg powder, flavours, egg white powder, lemon, baking agents: ammonium hydrogencarbonate and sodium hydrogencarbonate, emulsifier: soy lecithins Nutrients per 100 g: Fat 16 g, Carbohydrates 77 g, Protein 3,1 g 100 g/100g 98 g/100g Spiking Material Sample - 1,98 g/100g Table 2: Added amounts of lactose and fructose Ingredients Spiking material sample Sample B Potato flour 60,5 % 1,20 % Lactose 6,33 % 125 mg/100g Fructose 13,3 % 263 mg/100g * Drinking Powder, lactose free Ingredients: Sugar, maltodextrin, lactose free dry milk product, soluble coffee, deoilded cacoa powder, glucose syrup, hydrogenated vegetable fat, modified starch, salt, lactase, separating agent: E551, stabilizing agent: E415, E340, milk protein, flavour 20,0 % 20,0 % * Sample B contains additional fructose from the cookie matrix Page 4 of 45

5 2.1.1 Homogeneity Homogeneity of the test material was checked by 5fold determination of fructose by HPLC-RI. With a repeatability standard deviation of 5,8 % the homogeneity of the test material was considered acceptable. The precision data of similar methods for determination of mo- and disaccharides were for the relative repeatability standard deviation 0,46 2,9 % and for the relative reproducibility standard deviation 2,5 7,0 % for milk product containing foods (18). The results are given in the documentation. Additionally in the documentation the portion numbers are graphically assigned to the results of galactose (sample A) and lactose (sample B and spiking material sample). There is trend recognizable in the results which could suggest inhomogeneity. 2.2 Test Two portions of test material were sent to every participating laboratory in the 18 th week of The testing method was optional. The tests should be finished at 12 th june Submission of results The participants submitted their results in provided standard forms by . The final concentration of each parameter of the numbered samples was used for each statistical evaluation. Queried and documented were single results, recovery and the testing methods used. One laboratory did t submit results. All other participants submitted their results in time. Page 5 of 45

6 3. Evaluation 3.1 Assigned value Because the analysed material was certified reference material the robust mean of the submitted results was used as assigned value X (6). The distribution of submitted results showed hint for bimodal distribution or other reasons for a higher variability. The statistical evaluation is performed when there are at least 7 results for a parameter. Single results giving values outside the measuring range of the participating laboratory were considered for statistical evaluation (e.g. results given as > 25 mg/kg and < 2,5 mg/kg, respectively) when a result indicating > is above and a result indicating < is below the target range. 3.2 Standard deviation For comparison to the target standard deviation a robust standard deviation (S x ) was calculated (6). 3.3 Outliers Statistical outliers were determined by Mandel s-h-statistic for 95% significance niveau (5). Detected outliers were stated for information only, when z-score was < -2 or > Target standard deviation The target standard deviation of the assigned value is determined according to the following methods. In general the Horwitz target standard deviation is suitable for the statistical evaluation of interlaboratory tests where different analytical methods are applied. The standard deviation from precision experiments are derived from proficiency tests where a specific analytical method is mandatory. For information the z-scores from both models were given in the result tables, if available. Page 6 of 45

7 3.4.1 General model (Horwitz) The relative target standard deviation in % of the assigned value is derived from following equation (Horwitz) σ (%) = 2 (1-0,5logX) From the result the target standard deviation is calculated σ = X * σ (%) / Value by precision experiment Using the reproducibility standard deviation σ R and the repeatability standard deviation σ r of a precision experiment the between-laboratories standard deviation can be calculated σ L : L = R 2 r 2. And then, using the number of replicate measurements n, each participant is to perform, the target standard deviation for proficiency assessment is calculated : = L 2 r 2 /n. The target standard deviations given in table 3 were calculated from the precision data of the respective methods. Tabelle 3: Relative target standard deviations (RSD) from precision experiments 14, 15, 16, 17) Parameter Matrix RSD σ Method / Literature Fructose rusk 2,0 % UV-Test(14) Glucose rusk 1,9 % UV-Test(14) Lactose Lactose "lactose free" skimmed milk "lactose free" milk (3 samples) 24,5 % UV-Test (15) 9,2 % 3,7 % 4,9 % UV-Test (16) for determination in the presence of glucose Lactose milk 0,92 % HPLC-RI(17) Lactose cream 4,4 % HPLC-RI(17) Lactose milk powder 2,6 % HPLC-RI(17) Page 7 of 45

8 3.4.3 Value by perception The target standard deviation for proficiency assessment can be set at a value that corresponds to the level of performance that the coordinator would wish laboratories to be able to achieve (6). For the present evaluation of results the target standard deviation according to Horwitz was applied for fructose, lactose and galactose. Additionally the standard uncertainty was considered for fructose and lactose by valuating with z'-score (see 3.6). 3.5 z-score To assess the results of the participants the z-score is used. It indicates about which multiple of the target standard deviation ( σ ) the result (x) of the participant is deviating from the assigned value (X) (6). Participants z-scores were derived as: z = (x X) / σ ; the requirements for the analytical performance are generally considered as fulfilled if -2 z z'-score The z'-score can be used for the valuation of the results of the participants, in cases the standard uncertainty has to be considered (s. 3.8). The z'-score represents the relation of the deviation of the result (x) of the participant from the respective assigned value (X) to the square root of quadrat sum of the target standard deviation ( σ ) and the standard uncertainty (Ux) (6). Participants z'-scores are derived as: In the following we define the deminator deviation σ '. as the target standard The requirements for the analytical performance are generally considered as fulfilled if -2 z' 2. The z'-score was used for the analytes fructose and lactose (see also 3.7 and 3.8). Page 8 of 45

9 3.7 Quotient S x / Following the Horrat-value the results of a proficiency-test (PT) can be considered convincing, if the quotient of robust standard deviation and target standard deviation does t exceed the value of 2. A value > 2 means an insufficient precision, i.e. the analytical method is too variable, or the variation between the test participants is higher than estimated. Thus the comparability of the results is t given (11). In the present proficiency tests the quotients S x / σ were In the present proficiency test the quotient S x / σ for fructose and lactose was above 2,0. Therefore evaluation was done according to 3.6 by z'-scores. The resulting quotients S x / σ ' were < 2,0 for fructose and < 3,0 for lactose. For galactose satisfactory quotients S x / σ of < 2,0 were obtained. 3.8 Standard uncertainty The assigned value X has a standard uncertainty u X that depends on the analytical method, differences between the analytical methods used, the test material, the number of participant laboratories and perhaps on other factors. The standard uncertainty u X for this PT is calculated as follows (6). u X =1,25 S x / p If u X 0,3 the standard uncertainty of the assigned value needs t to be included in the interpretation of the results of the PT (6). Values exceeding 0,3 imply, that the target standard deviation could be too low with respect to the standard uncertainty of the assigned value. In the present proficiency tests the quotients U x / σ were between 0,59 and 0,73 for galactose and between 1,1 and 1,8 for fructose and lactose. At the same time for fructose and lactose the quotients S x / σ were > 2,0. Therefore the target range was extended according 3.6 z'-score considering the standard uncertainty for fructose and lactose. 3.9 Recovery rates: Spiking For the results of the spiking material sample and the spiked sample recovery rates were calculated by DLA with respect to the kwn content of added lactose. The related values of added lactose are given in 2.1 test material in table 2. As a range of acceptance RA for valuating participant's results the range of % for the recovery rates were deduced from published methods (15-19). For lactose results of the spiking material sample and the spiked sample recovery rates were calculated with respect to the kwn added content of lactose. The recovery rates were given for information only. No statistical evaluation was done. The recovery rates should exclusively give an estimation of the matrix- and/or processing influences. Page 9 of 45

10 4. Results All following tables are anymized. With the delivering of the evaluation-report the participants are informed about their individual evaluation-number. In the first table the characteristics are listed: Statistic Data Number of results Number of outliers Mean Median Robust mean (X) Robust standard deviation (S x ) Target range: Target standard deviation σ or σ ' Target standard deviation for information lower limit of target range (X - 2 σ ) or (X - 2 σ ') * upper limit of target range (X + 2 σ ) or (X + 2 σ ') * Quotient S x / σ ' Standard uncertainty u X Quotient u X / ' Number of results in the target range * Target range is calculated with z-score or z'-score In the second table the individual results of the participating laboratories are listed: Auswertenummer Parameter Evaluation [Einheit / Unit] number Abweichung [Einheit / Unit] Deviation [Einheit / Unit] z-score σ z-score (Info) Hinweis Remark Page 10 of 45

11 4.1 Fructose Fructose Sample A (in mg/100g) Vergleichsuntersuchung / Proficiency Test Statistic Data Number of results 8 Number of outliers 1 Mean 897 Median 854 Robust Mean (X) 874 Robust standard deviation (S x ) 90,5 Target range: Target standard deviation σ ' 53,6 Target standard deviation Horwitz (for Information) 35,7 lower limit of target range 766 upper limit of target range 981 Quotient S x / σ ' Standard uncertainty u x Quotient u x / σ ' 1,7 40,0 0,75 Results in the target range 7 Percent in the target range 88 Ergebnisse / Results Auswertenummer / evaluation number Fructose [mg/100g] obere Grenze upper limit rob. Mittelwert robust mean untere Grenze lower limit Abb. 1: Ergebnisse Probe A Fig. 1: Results Sample A Page 11 of 45

12 Ergebnisse der teilnehmenden Institute: Results of Participants: Auswertenummer Evaluation number Fructose [mg/100g] Abweichung [mg/100g] Deviation [mg/100g] z'-score σ z-score (Info) ,51-1,5-2, ,49 0,4 0, ,51-1,3-2, ,51-1,2-1, ,51-0,5-0, ,49 1,8 2, ,49 6,1 9, ,51-0,3-0,4 Hinweis Remark Ausreisser / Outlier 3 z'-score > Abb. 2: Fig. 2: Auswertenummer / evaluation number Z'-Scores Probe A Z'-Scores Sample A Page 12 of 45

13 4.1.2 Fructose Sample B (in mg/100g) Vergleichsuntersuchung / Proficiency Test Statistic Data Number of results 8 Number of outliers 1 Mean 1153 Median 1114 Robust Mean (X) 1129 Robust standard deviation (S x ) 122 Target range: Target standard deviation σ ' 69,7 Target standard deviation Horwitz (for Information) 44,3 lower limit of target range 990 upper limit of target range 1268 Quotient S x / σ ' Standard uncertainty u x Quotient u x / σ ' 1,7 53,8 0,77 Results in the target range 7 Percent in the target range 88 Ergebnisse / Results Auswertenummer / evaluation number Fructose [mg/100g] obere Grenze upper limit rob. Mittelwert robust mean untere Grenze lower limit Abb. 3: Ergebnisse Probe B Fig. 3: Results Sample B Page 13 of 45

14 Ergebnisse der teilnehmenden Institute: Results of Participants: Auswertenummer Evaluation number Fructose [mg/100g] Abweichung [mg/100g] Deviation [mg/100g] z'-score σ z-score (Info) ,6-2, ,3 0, ,8-2, ,8-1, ,4 0, ,6 2, ,3 8, ,7-1,1 Hinweis Remark Ausreisser / Outlier 3 z'-score > Abb. 4: Fig. 4: Auswertenummer / evaluation number Z'-Scores Probe B Z'-Scores Sample B Page 14 of 45

15 4.1.3 Fructose Spiking Material Sample (in g/100g) Vergleichsuntersuchung / Proficiency Test Statistic Data Number of results 9 Number of outliers 1 Mean 13,6 Median 13,4 Robust Mean (X) 13,6 Robust standard deviation (S x ) 0,930 Target range: Target standard deviation σ ' 0,534 Target standard deviation Horwitz (for Information) 0,368 lower limit of target range 12,6 upper limit of target range 14,7 Quotient S x / σ ' Standard uncertainty u x Quotient u x / σ ' 1,7 0,39 0,73 Results in the target range 7 Percent in the target range 78 Ergebnisse / Results Auswertenummer / evaluation number Fructose [g/100g] obere Grenze upper limit rob. Mittelwert robust mean untere Grenze lower limit Abb. 5: Ergebnisse Dotierungsmaterialprobe Fig. 5: Results Spiking Material Sample Page 15 of 45

16 Ergebnisse der teilnehmenden Institute: Results of Participants: Auswerte- Fructose [g/100g] Abweichung z'-score z-score Hinweis nummer [g/100g] σ (Info) Evaluation Deviation Remark number [g/100g] 1 13,1-0,53-1,0-1,4 2 13,26-0,37-0,7-1,0 3 13,7 0,07 0,1 0,2 4 11,9-1,73-3,2-4, ,37-0,26-0,5-0, ,94 0,31 0,6 0, ,61 0,98 1,8 2, ,0 1,37 2,6 3, ,4-0,23-0,4-0,6 Ausreisser / Outlier Einheit korrigiert / units corrected 4 z'-score Abb. 6: Fig. 6: Auswertenummer / evaluation number Z'-Scores Dotierungsmaterialprobe Z'-Scores Spiking Material Sample Page 16 of 45

17 4.2 Lactose Qualitative Valuation Sample A and Sample B Vergleichsuntersuchung / Proficiency Test Evaluation number Sample A pos/neg Sample A Sample B Sample B [m g/kg] pos/neg [m g/kg] Qualitative Valuation Agreement with Consensus Value Remarks 1 positive < 10 positive 129 1/1 (100%) 2 negative < 10 positive 121 1/1 (100%) 3 positive 4,04 positive 134 1/1 (100%) 4 positive < 50 positive 145 1/1 (100%) 5 negative < 50 positive 62 1/1 (100%) 6 negative < 0,04 positive 0,125 1/1 (100%) 7 positive 16 positive 271 1/1 (100%) 8 positive < 5 positive 96,4 1/1 (100%) 9 positive n.b. positive 128 1/1 (100%) Sample A: 13 mg/100g 10 negative positive 115 1/1 (100%) 11 negative n.n. positive 125 1/1 (100%) 12 positive < BG positive 146 1/1 (100%) 13 positive < 10 positive 146 1/1 (100%) 14 positive 7,39 positive 17,291 1/1 (100%) 15 negative < 10 positive 100 1/1 (100%) 16 positive < 10 positive 120 1/1 (100%) Number positive Number negative Percent positive Percent negative Consensus Sample A Sample B ne 0 positive Anmerkung: There were 100% positive results for sample B. For sample A consensus value of > 75% was obtained. For sample A all results were < 50 mg/100g and thereof 8 results < 10 mg/100g. 38% of the participants stated lactose was detectable in sample A. Page 17 of 45

18 4.2.2 Lactose Sample B (in mg/100g) Vergleichsuntersuchung / Proficiency Test Statistic Data Number of results 16 Number of outliers 2 Mean 116 Median 123 Robust Mean (X) 116 Robust standard deviation (S x ) 37,3 Target range: Target standard deviation σ ' 13,3 Target standard deviation (for Information) 6,43 lower limit of target range 89,6 upper limit of target range 143 Quotient S x / σ ' Standard uncertainty u x Quotient u x / σ ' 2,8 11,7 0,88 Results in the target range 9 Percent in the target range 56 Ergebnisse / Results > Auswertenummer / evaluation number Lactose [mg/100g] obere Grenze upper limit rob. Mittelwert robust mean untere Grenze lower limit Abb. 7: Ergebnisse Probe B Fig. 7: Results Sample B Page 18 of 45

19 Ergebnisse der teilnehmenden Institute: Results of Participants: Auswerte- Lactose [mg/100g] Abweichung z'-score z-score Hinweis nummer [mg/100g] σ (Info) Evaluation Deviation Remark number [mg/100g] ,71 1,0 2, ,71 0,4 0, ,71 1,3 2, ,71 2,2 4, ,29-4,1-8,4 6 0, ,16-8,7-18, ,71 11,6 24,1 8 96,4-19,89-1,5-3, ,71 0,9 1, ,29-0,1-0, ,71 0,7 1, ,71 2,2 4, ,71 2,2 4, , ,4-15, ,29-1,2-2, ,71 0,3 0,6 Ausreisser / Outlier; Einheitenfehler / unit error? Ausreisser / Outlier z'-score > < < Auswertenummer / evaluation number Abb. 10: Fig. 10: Z'-Scores Probe B Z'-Scores Sample B Page 19 of 45

20 4.2.3 Lactose Spiking Material Sample (in g/100g) Vergleichsuntersuchung / Proficiency Test Statistic Data Number of results 15 Number of outliers 2 Mean 5,69 Median 6,22 Robust Mean (X) 5,96 Robust standard deviation (S x ) 0,642 Target range: Target standard deviation σ ' 0,276 Target standard deviation Horwitz (for Information) 0,182 lower limit of target range 5,41 upper limit of target range 6,52 Quotient S x / σ ' Standard uncertainty u x Quotient u x / σ ' 2,3 0,21 0,75 Results in the target range 10 Percent in the target range 67 Ergebnisse / Results Auswertenummer / evaluation number Lactose [g/100g] obere Grenze upper limit rob. Mittelwert robust mean untere Grenze lower limit Abb. 11: Ergebnisse Dotierungsmaterialprobe Fig. 11: Results Spiking Material Sample Page 20 of 45

21 Ergebnisse der teilnehmenden Institute: Results of Participants: Auswerte- Lactose [g/100g] Abweichung z'-score z-score Hinweis nummer [g/100g] σ (Info) Evaluation Deviation Remark number [g/100g] 1 6,4 0,44 1,6 2,4 2 6,4 0,44 1,6 2,4 3 3,09-2,87-10,4-15,8 4 6,28 0,32 1,1 1,7 5 2,8-3,16-11,5-17, ,227 0,26 1,0 1,4 8 5,75-0,21-0,8-1,2 9 6,219 0,26 0,9 1,4 10 5,7-0,26-1,0-1,4 11 6,01 0,05 0,2 0,3 12 5,28-0,68-2,5-3,7 13 6,57 0,61 2,2 3,3 14 5,48-0,48-1,8-2,7 15 6,80 0,84 3,0 4,6 16 6,33 0,37 1,3 2,0 Ausreisser / Outlier Einheit korrigiert / units corrected 4 z'-score < < Auswertenummer / evaluation number Abb. 12: Fig. 12: Z'-Scores Dotierungsmaterialprobe Z'-Scores Spiking Material Sample Page 21 of 45

22 4.2.4 Recovery Rates for Lactose Hereafter the recovery rates of the participants' results with respect to the level of addition (page 4, table 2) were calculated by DLA and given for information only. Spiking Material Sample and Sample B Evaluation number Spiking material Recovery rate Sample B Recovery rate [g/100g] [%] [m g/100g] [%] 1 6, , , , , ,125 0,10 7 6, , , , , , , , , , , , Remarks RA* % RA* % 12 Anzahl im AB 7 Number in RA Percent in RA 80 Prozent im AB 44 * Range of acceptance see 3.9 Anmerkung: 80% of the participants obtained a recovery rate within the range of %. For the spiked sample B produced with the spiking material sample 44% of the recovery rates were in this range. Considering possible lactose content of the matrix at the limits of determination of the methods (see sample A), the level of lactose could be appr mg increased. Therefore the recovery rates of the participants 4, 12 and 13 for sample B are acceptable too. Page 22 of 45

23 4.3 Galactose Galactose Sample A (in mg/100g) Vergleichsuntersuchung / Proficiency Test Statistic Data Number of results 10 Number of outliers 1 Mean 274 Median 242 Robust Mean (X) 245 Robust standard deviation (S x ) 18,2 Target range: Target standard deviation σ 12,1 lower limit of target range 220 upper limit of target range 269 Quotient S x / σ Standard uncertainty u x Quotient u x / σ 1,5 7,2 0,59 Results in the target range 8 Percent in the target range 80 Ergebnisse / Results Auswertenummer / evaluation number Galactose [mg/100g] obere Grenze upper limit rob. Mittelwert robust mean untere Grenze lower limit Abb. 13: Ergebnisse Probe A Fig. 13: Results Sample A Page 23 of 45

24 Ergebnisse der teilnehmenden Institute: Results of Participants: Auswertenummer Evaluation number Galactose [mg/100g] Abweichung [mg/100g] Deviation [mg/100g] z-score σ ,58-1, ,58 0, ,42 26, ,42 0, ,58-0, ,58-1, ,58-1, ,42 0, ,58-0, ,42 2,1 Hinweis Remark Ausreisser / Outlier 3 z-score > Auswertenummer / evaluation number Abb. 14: Fig. 14: Z-Scores Probe A Z-Scores Sample A Page 24 of 45

25 4.3.2 Galactose Sample B (in mg/100g) Vergleichsuntersuchung / Proficiency Test Statistic Data Number of results 10 Number of outliers 1 Mean 266 Median 247 Robust Mean (X) 249 Robust standard deviation (S x ) 21,8 Target range: Target standard deviation σ 12,3 lower limit of target range 224 upper limit of target range 274 Quotient S x / σ Standard uncertainty u x Quotient u x / σ 1,8 8,6 0,70 Results in the target range 8 Percent in the target range 80 Ergebnisse / Results Auswertenummer / evaluation number Galactose [mg/100g] obere Grenze upper limit rob. Mittelwert robust mean untere Grenze lower limit Abb. 15: Ergebnisse Probe B Fig. 15: Results Sample B Page 25 of 45

26 Ergebnisse der teilnehmenden Institute: Results of Participants: Auswertenummer Evaluation number Galactose [mg/100g] Abweichung [mg/100g] Deviation [mg/100g] z-score σ , , , , , , , , , ,5 Hinweis Remark Ausreisser / Outlier 3 z-score > Auswertenummer / evaluation number Abb. 16: Fig. 16: Z-Scores Probe B Z-Scores Sample B Page 26 of 45

27 4.3.3 Galactose Spiking Material Sample (in g/100g) Vergleichsuntersuchung / Proficiency Test Statistic Data Number of results 11 Number of outliers 1 Mean 0,505 Median 0,470 Robust Mean (X) 0,478 Robust standard deviation (S x ) 0,0352 Target range: Target standard deviation σ 0,0214 lower limit of target range 0,436 upper limit of target range 0,521 Quotient S x / σ Standard uncertainty u x Quotient u x / σ 1,6 0,01 0,62 Results in the target range 9 Percent in the target range 82 Ergebnisse / Results 1,00 0,90 0,80 0,70 0,60 0,50 0,40 0,30 0,20 0,10 0, Auswertenummer / evaluation number Galactose [g/100g] obere Grenze upper limit rob. Mittelwert robust mean untere Grenze lower limit Abb. 17: Ergebnisse Dotierungsmaterialprobe Fig. 17: Results Spiking Material Sample Page 27 of 45

28 Ergebnisse der teilnehmenden Institute: Results of Participants: Auswertenummer Evaluation number Galactose [g/100g] Abweichung [g/100g] Deviation [g/100g] z-score σ 1 2 0,46-0,018-0,9 3 0,46-0,018-0,9 4 0,5 0,022 1,0 5 0,84 0,362 16, ,414-0,064-3, ,506 0,028 1,3 10 0,45-0,028-1,3 11 0,46-0,018-0,9 12 0,5 0,022 1, ,470-0,008-0,4 16 0,5 0,022 1,0 Hinweis Remark Ausreisser / Outlier Einheit korrigiert / units corrected z-score > Auswertenummer / evaluation number Abb. 18: Fig. 18: Z-Scores Dotierungsmaterialprobe Z-Scores Spiking Material Sample Page 28 of 45

29 5. Documentation 5.1 Primary data Parameter Teilnehmer Einheit Nachweisbar NWG BG Ergebnis Inkl. WF Wiederfindungsrate [%] WF mit gleicher Matrix Analyte Participant Unit Detectable LOD LOQ Result Incl. RR Recovery rate [%] RR with same Matrix 1 mg/100g yes mg/100g yes mg/100g 4 mg/100g yes mg/100g t partic. 6 mg/100g 7 mg/100g - - Fructose 8 mg/100g Probe A / 9 mg/100g yes 810 Sample A 10 mg/100g 11 mg/100g 12 mg/100g yes mg/100g mg/100g 15 mg/100g yes mg/100g yes n/a n/a Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 29 von 45

30 Parameter Teilnehmer Einheit Nachweisbar NWG BG Ergebnis Inkl. WF Wiederfindungsrate [%] WF mit gleicher Matrix Analyte Participant Unit Detectable LOD LOQ Result Incl. RR Recovery rate [%] RR with same Matrix 1 mg/100g yes mg/100g yes mg/100g 4 mg/100g yes mg/100g t partic. 6 mg/100g 7 mg/100g - - Fructose 8 mg/100g Probe B / 9 mg/100g yes 1070 Sample B 10 mg/100g 11 mg/100g 12 mg/100g yes mg/100g mg/100g 15 mg/100g yes mg/100g yes n/a n/a Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 30 von 45

31 Parameter Teilnehmer Einheit Nachweisbar NWG BG Ergebnis Inkl. WF Wiederfindungsrate [%] WF mit gleicher Matrix Analyte Participant Unit Detectable LOD LOQ Result Incl. RR Recovery rate [%] RR with same Matrix 1 g/100g yes 0,1 13,1 2 g/100g yes 0,01 13, g/100g yes 0,01 0,01 13,7 4 g/100g yes 0,05 0,1 11,9 5 g/100g 6 g/100g Fructose 7 g/100g - - Dotierungsmaterialprobe / 8 g/100g 9 g/100g yes 13,37 Spiking Material Sample 10 g/100g 11 g/100g 12 g/100g yes ,94 13 g/100g 0,5 14,61 14 g/100g 15 g/100g yes g/100g yes n/a 0,01 13,4 n/a Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 31 von 45

32 Parameter Teilnehmer Einheit Nachweisbar NWG BG Ergebnis Inkl. WF Wiederfindungsrate [%] WF mit gleicher Matrix Analyte Participant Unit Detectable LOD LOQ Result Incl. RR Recovery rate [%] RR with same Matrix 1 mg/100g yes 1 10 < ,4 2 mg/100g 10 < mg/100g yes 1 2 4,04 yes 88 yes 4 mg/100g yes <50 5 mg/100g 0,01 0,05 <50 yes 95 yes t 6 mg/100g determined, t because t determined relevant for 0,04 < 0, yes valuation 7 mg/100g yes 16 8 mg/100g yes 5 mg/100g <5 Lactose n.b. (13 9 mg/100g yes 4 14 Probe A / mg/100g) Sample A No (the recovery was calculated from the concentrtion of 10 mg/100g 8 mg/100g 99 measured lactose in the standard solution compared to the concentration lactose written on the bottle 11 mg/100g 70 n.n. 99,8 12 mg/100g yes 3 10 <BG 13 mg/100g yes 10 <10 14 mg/100g yes 3mg / 100g 3mg / 100g 7,39 15 mg/100g 5 10 <10-16 mg/100g yes n/a 10 <10 n/a Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 32 von 45

33 Parameter Teilnehmer Einheit Nachweisbar NWG BG Ergebnis Inkl. WF Wiederfindungsrate [%] WF mit gleicher Matrix Analyte Participant Unit Detectable LOD LOQ Result Incl. RR Recovery rate [%] RR with same Matrix 1 mg/100g yes ,4 2 mg/100g yes mg/100g yes yes 62 yes 4 mg/100g yes mg/100g yes 0,01 0,05 62 yes 95 yes 6 mg/100g yes t determined, because t relevant for 0,04 0, yes Lactose valuation Probe B / 7 mg/100g yes 271 Sample B 8 mg/100g yes 5 mg/100g 96,4 9 mg/100g yes mg/100g 8 mg/100g mg/100g yes ,8 12 mg/100g yes mg/100g mg/100g yes 3mg / 100g 3mg / 100g 17,29 15 mg/100g yes mg/100g yes n/a n/a Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 33 von 45

34 Parameter Teilnehmer Einheit Nachweisbar NWG BG Ergebnis Inkl. WF Wiederfindungsrate [%] WF mit gleicher Matrix Analyte Participant Unit Detectable LOD LOQ Result Incl. RR Recovery rate [%] RR with same Matrix 1 g/100g yes 0 0,01 6,4 101,4 2 g/100g yes 0,01 6, g/100g yes 0,01 0,01 3,09 4 g/100g yes 0,05 0,1 6,28 5 g/100g yes 0,01 0,05 2,8 yes 95 yes 6 g/100g Lactose 7 g/100g yes 6,227 Dotierungsmaterialprobe 8 g/100g yes 5 mg/100g 5,75 / 9 g/100g yes 0 0,01 6,219 Spiking Material 10 g/100g 8 mg/100g 5.7 Sample 11 g/100g yes 0,07 6,01 94,7 12 g/100g yes ,28 13 g/100g 0,5 6,57 14 g/100g yes 0.003g / 0.003g / 100g 100g 5,48 15 g/100g yes g/100g yes n/a 0,01 6,33 n/a Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 34 von 45

35 Parameter Teilnehmer Einheit Nachweisbar NWG BG Ergebnis Inkl. WF Wiederfindungsrate [%] WF mit gleicher Matrix Analyte Participant Unit Detectable LOD LOQ Result Incl. RR Recovery rate [%] RR with same Matrix 1 mg/100g 2 mg/100g yes mg/100g 4 mg/100g yes mg/100g yes 0,01 0, yes 95 yes 6 mg/100g 7 mg/100g yes 251 Galactose 8 mg/100g Probe A / 9 mg/100g yes 239 Sample A 10 mg/100g 8 mg/100g mg/100g yes ,8 12 mg/100g yes mg/100g 14 mg/100g 15 mg/100g yes mg/100g yes n/a n/a Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 35 von 45

36 Parameter Teilnehmer Einheit Nachweisbar NWG BG Ergebnis Inkl. WF Wiederfindungsrate [%] WF mit gleicher Matrix Analyte Participant Unit Detectable LOD LOQ Result Incl. RR Recovery rate [%] RR with same Matrix 1 mg/100g 2 mg/100g yes mg/100g 4 mg/100g yes mg/100g yes 0,01 0, yes 95 yes 6 mg/100g 7 mg/100g yes 227 Galactose 8 mg/100g Probe B / 9 mg/100g yes 240 Sample B 10 mg/100g 8 mg/100g mg/100g yes ,8 12 mg/100g yes mg/100g 14 mg/100g 15 mg/100g yes mg/100g yes n/a n/a Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 36 von 45

37 Parameter Teilnehmer Einheit Nachweisbar NWG BG Ergebnis Inkl. WF Wiederfindungsrate [%] WF mit gleicher Matrix Analyte Participant Unit Detectable LOD LOQ Result Incl. RR Recovery rate [%] RR with same Matrix 1 g/100g 2 g/100g yes 0,01 0, g/100g yes 0,01 0,01 0,46 4 g/100g yes 0,05 0,1 0,5 5 g/100g yes 0,01 0,05 0,84 yes 95 yes 6 g/100g Galactose 7 g/100g yes 0,41 Dotierungsmaterialprobe / 8 g/100g Spiking Material 9 g/100g yes 0,51 Sample 10 g/100g 8 mg/100g g/100g yes 0,04 0,46 94,7 12 g/100g yes ,5 13 g/100g 14 g/100g 15 g/100g yes g/100g yes n/a 0,01 0,5 n/a Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 37 von 45

38 5.2 Homogeneity Homogeneity test before the PT Homogeneity test for fructose by HPLC-RI: Independant samples g/100g 1 0,88 2 0,86 3 0,93 4 0,81 5 0,83 Mean Repeatability Standard Deviation 0,86 0,047 5,4% Comparison of sample number / test results 120 Homgenität / homogeneity Galactose Sample A DLA-Nr. / No. Ergebnis : Page 38 of 45

39 Homgenität / homogeneity Lactose Sample B DLA-Nr. / No. Ergebnis : Homgenität / homogeneity Lactose Dotierungsmaterialprobe DLA-Nr. / No. Ergebnis *10 Page 39 of 45

40 5.3 Analytical Methods Details by the participants Teilnehmer Analyten Methode ist akkreditiert Methodenbeschreibung Probenvorbereitung Einwaage Extraktion Referenzmaterial Sonstige Hinweise Participant Analytes Method accredited Method description Sample Preparation Sample weight Extraction Referencematerial Further remarks 1 Fructose yes enzymatically 2,5 2 Fructose yes SLMB Detektor: Amperometric Cell dissolve in H2O 1g 40min/70 C 3 Fructose yes IC-PAD dissolved sample in water 1 g water Sigma 4 Fructose yes Microgenics Test- Combination Fructose 6 Fructose 7 Fructose Fructose 64 LFBG, , December 2002 modified addition of hot water (enzymatic determination by 9 Fructose yes ne appr. 1 g (60 C-70 C), 15 min D (-)-Fructose, Sigma magnetic stirrer 10 Fructose 11 Fructose 12 Fructose test kit Saccharose/D- Glucose/D-Fructose, Fa. R- biopharm) yes enzymatically homogenization 1g/10ml 1+2 ; 3g/100ml 3 aquaeous Extraction r-biopharm 13 Fructose 5g HPLC-Refractive index Water extraction 5g 1+2 ; 2,5 g ,5 g 3 diluted in 25 ml 14 Fructose 15 Fructose yes HPIEC-PAD Water extraction 1 g / 100 ml 60 C; Ultraschall Fluka 16 Fructose hplc extracted in water 5g 1+2 ; 2,5g Probe A und B / Sample A and B 3 Dotierungsmaterialprobe / Spiking Material Sample LOD and LOQ t determined, because t relevant for routine analysis Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 40 von 45

41 Teilnehmer Analyten Methode ist akkreditiert Methodenbeschreibung Probenvorbereitung Einwaage Extraktion Referenzmaterial Sonstige Hinweise Participant Analytes Method accredited Method description Sample Preparation Sample weight Extraction Referencematerial Further remarks 1 Lactose enzymatically 2 2 Lactose yes SLMB Detektor: Amperometric Cell dissolve in H2O 1g 40min/70 C recovery calculated 3 Lactose 1+2 HPLC-MS shaking 1 g formate buffer Fluka by C13-Lactose internal standard 3 Lactose 3 yes IC-PAD dissolved sample 1 g water Sigma anhydrous lactose Lactose-Mohydrate, Merck r-biopharm Test- 4 Lactose yes Combination Lactose yes Enzymatically boehringer water 2g 70 c - ASU L Analysis of foods extraction with hot Determination of lactose water 70 C, Carrezclarification, Glucose- content in lactose-reduced 6 Lactose yes milk and milk products in 2,5 g elimination by Enzytec the presence of glucose Glucose-Remover-Kit enzymatic method, E3400 modification: application of a test kit 7 Lactose yes enzymatically Carrez-clarification 15 Min, 70 C DRRR, DLA 8 Lactose yes AOAC mod /IC-EC 64 LFGB, L , Sept modified (enzymatic 9 Lactose yes determination by test kit ne ca. 1 g Lactose/D- Galactose, Fa. r-biopharm) water bath 70 C, 15 min standard solutions from test kits 1+2 Probe A und B / Sample A and B 3 Dotierungsmaterialprobe / Spiking Material Sample extraction with hot water 70 C, Carrezclarification Lactosedetermination by Boehringer Mannheim/R- Biopharm Lactose/D- Glucose-Enzyme test kit , resiöts as anhydrous lactose Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 41 von 45

42 Teilnehmer Analyten Methode ist akkreditiert Methodenbeschreibung Probenvorbereitung Einwaage Extraktion Referenzmaterial Sonstige Hinweise Participant Analytes Method accredited Method description Sample Preparation Sample weight Extraction Referencematerial Further remarks 60ml MQ water was added and the sample 10 Lactose yes was incubated in The method is t water bath (70C) for suitable for analysis 15 min. After cooling, NMKL metod No 155, 2nd of lactose in samples 5 ml Carrez I, 5ml ed 2006, SLV -m045.1 that contain more The sample was Carrez II and 10ml In house reference which includes analysis with than 5 times more shaken and 2.0 g were 2.0 g 0.1M NaOH was material made from the testkit Lactose/Dgalctose UV-Test from R- to lactose and/or galactose compared taken for analysis. added and the extract cornflakes was incubated for 30 biopharm samples that have min. Water was added been pretreated with to a total volume of lactase. 100 ml. The sample was centrifuged to remove e.g. fat. 11 Lactose yes Enzymatically; UV 340 nm Carrez-clarification 2,42 15 min, 60 C 12 Lactose 1+2 HPAEC-PAD homogenisation 1g/10ml aqueous Extraction 12 Lactose 3 yes Enzymatically homogenisation 3g/100ml aqueous Extraction r-biopharm Ion chromatography- 13 Lactose 1+2 Amperometric pulse Water extraction 0,25 g 0,25 g diluted in 25 ml detector 13 Lactose 3 HPLC-Refractive index Water extraction 2,5 g 2,5 g diluted in 25 ml 14 Lactose Enzymatic (Megazyme Kit) Water Extraction & Carrex Clarification 1g 50'c for 30 minutes In House Spike 15 Lactose yes HPIEC-PAD Water Extraction 1 g / 100 ml 60 C; Ultraschall Fluka 16 Lactose hplc extracted in water 5g 1+2 Probe A und B / Sample A and B 3 Dotierungsmaterialprobe / Spiking Material Sample Megazyme Kit K- LACGAR Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 42 von 45

43 Teilnehmer Analyten Methode ist akkreditiert Methodenbeschreibung Probenvorbereitung Einwaage Extraktion Referenzmaterial Sonstige Hinweise Participant Analytes Method accredited Method description Sample Preparation Sample weight Extraction Referencematerial Further remarks 1 Galactose 2 Galactose yes SLMB Detector: Amperometric Cell dissolve in H2O 1g 40min/70 C 3 Galactose r-biopharm Test- 4 Galactose yes Combination Galactose yes Enzymatik boehringer 6 Galactose i 7 Galactose yes enzymatically Carrez-clarification 15 Min, 70 C DLA (08/2014) 8 Galactose 64 LFGB, L , Sept modified (enzymatic 9 Galactose yes determination test kit keine ca. 1 g Lactose/D- Galactose, Fa. r-biopharm) water bath 70 C, 15 min reference material, LOD, because t relevant for routine analysis 10 Galactose yes See above 11 Galactose yes Enzymatically; UV 340 nm Carrez-clarification 2,421 / 2, ; 5, min, 60 C Galactose yes Enzymatically homogenisation Galactose Galactose 15 Galactose yes HPIEC-PAD Extraction with water 1 g / 100 ml 60 C; Ultrasonic bath Sigma-Aldrich 16 Galactose hplc extracted in water 5g 1g/10ml 1+2 ; 3g/100ml 3 aqueous Extraction r-biopharm 1+2 Probe A und B / Sample A and B 3 Dotierungsmaterialprobe / Spiking Material Sample Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 43 von 45

44 6. Index of participant laboratories in alphabetical order Teilnehmer / Participant Ort / Town Land / Country GREAT BRITAIN AUSTRIA FRANCE Germany ITALY Germany SPAIN Germany DENMARK Germany Germany Germany Germany SWEDEN GREAT BRITAIN Germany Germany [The address data of the participants were deleted for publication of the evaluation report.] Nachdruck, auch auszugsweise, nur mit schriftlicher Genehmigung von DLA-Ahrensburg Seite 44 von 45

45 7. Index of references 1. DIN EN ISO/IEC 17043:2010; Konformitätsbewertung Allgemeine Anforderungen an Eignungsprüfungen / Conformity assessment General requirements for proficiency testing 2. Verordnung / Regulation 882/2004/EU; Verordnung über amtliche Kontrollen / Regulation on official controls 3. DIN EN ISO/IEC 17025:2005; Allgemeine Anforderungen an die Kompetenz von Prüf- und Kalibrierlaboratorien / General requirements for the competence of testing and calibration laboratories 4. Richtlinie / Directive 1993/99/EU; über zusätzliche Maßnahmen im Bereich der amtlichen Lebensmittelüberwachung / on additional measures concerning the official control of foodstuffs 5. ASU 64 LFGB : Planung und statistische Auswertung von Ringversuchen zur Methodenvalidierung 6. DIN ISO 13528:2009; Statistische Verfahren für Eignungsprüfungen durch Ringversuche / Statistical methods for use in proficiency testing by interlaboratory comparisons 7. The International Harmonised Protocol for the Proficiency Testing of Ananlytical Laboratories ; J.AOAC Int., 76(4), (1993) 8. The International Harmonised Protocol for the Proficiency Testing of Ananlytical Chemistry Laboratories ; Pure Appl Chem, 78, (2006) 9. Evaluation of analytical methods used for regulation of food and drugs;w. Horwitz; Analytical Chemistry, 54, (1982) 10.A Horwitz-like funktion describes precision in proficiency test; M. Thompson, P.J. Lowthian; Analyst, 120, (1995) 11.Protocol for the design, conduct and interpretation of method performance studies; W. Horwitz; Pure & Applied Chemistry, 67, (1995) 12.Recent trends in inter-laboratory precision at ppb and sub-ppb concentrations in relation to fitness for purpose criteria in proficiency testing; M. Thompson; Analyst, 125, (2000) 13.ASU 64 LFGB L (2010) / DIN : Bestimmung des Lactose- und Galactosegehaltes von Milch und Milchprodukten; Enzymatisches Verfahren / Milk and milk products - Determination of lactose and D-galactose content - Enzymatic method 14.ASU 64 LFGB L Bestimmung des Lactosegehaltes in lactosereduzierter Milch und lactosereduzierten Milchprodukten in Gegenwart von Glucose; Enzymatisches Verfahren (2014) 15.ASU 64 LFGB L Bestimmung von Lactose in Brot einschließlich Kleingebäck aus Brotteigen (1983) 16.ASU 64 LFGB L Bestimmung von Glucose und Fructose in Kinder- Zwieback und Zwiebackmehl (1985). 17.ISO 22662:2012; Milch und Milchprodukte - Bestimmung des Lactosegehalts mit Hochleistungs-Flüssigchromatographie (Referenzverfahren) / Milk and milk products Determination of lactose content by high-performance liquid chromatography (Reference method) 18.Chávez-Servín et al. (2004) Analysis of mo- and disaccharides in milkbased formulae by high-performance liquid chromatography with refractive index detection. J Chromatogr A. 1043(2): Morlock et al. (2014) Streamlined analysis of lactose-free dairy products. Chromatogr A. 1324: Page 45 of 45