435 / 696
Validation Results
Limit of Detection (LOD)
The Limit of Detection for the analysis was determined by LOD = ((3 x standard
deviation of blank) + mean of blank). Ten blank replicates were utilized to calculate the
LOD. An instrumental LOD of 680 pg for carnitine and 940 pg for choline was obtained.
This equates to an LOD of 0.0034 mg/100g for carnitine and 0.0047 mg/100g for choline
by this method when utilizing the default weights and dilution factors. This is well below
the required LOD of both SMPR.
Linearity
The linearity was assessed by the percent deviation from theoretical concentration across
the calibration (Table 4). The calibration range was from 10 ng/mL to 5000 ng/mL. A
linear model with a weighting of 1/x
2
was used to best represent the response across this
range. The correlation coefficient across these same three analyses is included in Table
5. Both tests show acceptable linearity across the calibration range.
Table 4. Deviation from Theoretical, n = 3 analyses
Conc. (ng/mL)
10
20
500
2000
4000
5000
Analyte
Percent Deviation from Theoretical
Mean
Carnitine
2.9
3.0
2.0
2.4
1.2
2.0
2.2
Choline
2.5
4.1
1.4
1.1
1.0
1.7
1.9
Table 5. Correlation Coefficient (r), n = 3 analyses
Analyte
Mean Correlation Coefficient
Carnitine
0.9994
Choline
0.9993
Specificity
Specificity was determined through equivalence of the ion ratios across two different
transitions in both the free and total analyses for all non-placebo materials (Tables 6-9).
The comparison of the ion ratio between sample and reference standards showed %
across all matrices, with the majority under 5%. This method can not determine between
D and L-carnitine. The majority of these products are fortified with L-carnitine, and all
concentrations were below quantitation levels in the placebo products. There was really
no considerable increases from free to total analysis except in the milk based products,
which supports the presence of bound forms of carnitine.
Chol-08/Carn-07 (February 2016)
FOR ERP USE ONLY
DO NOT DISTRIBUTE