460-461 / 620
This method (MTE-02) is the same MW-ICP-MS method as CHROM-06 (2011.19) with the additional nine elements added and determined simultaneously on the same digest.
Table 1
Spike recoveries for all 12 elements in 90-110% range as required. Levels in red are below PLOQ.
Table 2a-2l
Table 3
Table 4
Table 5a
Table 5b
Table 6
2011.19 is already Final Action and has completed MLT study. As part of that study, labs were invited to submit data on the other elements if they wanted to. Six labs submitted,with one of those
omitting the major elements.
Table 6 contains data relating to the accuracy of the method: a comparison of MLT results to ICP-AES, from a 6day x 2 SLV study of Abbott's internal MW-ICP-AES method (similar,
no doubt, to MTE-01). In general, ICP-MS results are 2-5% higher than ICP-AES results, except for the SRM which is very close.
In a nutshell, the 12-element method performed as well as the ultratrace method 2011.19 and met almost all of the SMPR criteria. The only main issues with the method are the currently high results
being obtained for P on the SRM 1849a, high RSD (R) for P, and the fact that performance was not tested at the low levels of Mn required by the SMPR (may or may not be a problem). The method MTE-
02 is currently not written to support determinations of very low Mn becuase the calibration curve and linearity tests stop at 5 ppb in analytical solution, which with a 10x dilution would be 50 ng/g in
product, or 0.005 mg/100g (but the SMPR requires 0.001 mg/100g).
SRM statistical analysis. The values entered are from our latest control charts. Some elements are above or below the certified limits, but not by much percentage-wise - see
Method Evaluation Form for summary. The big issue is P, which is 8% higher than certified mean. However, this is not consistent with earlier validation work done with this
method, nor with results obtained during the MLT (Table 6). Indeed, the MLT results for P were only slightly above certified value and the other SPIFAN materials agreed well with
ICP-AES results from Abbott for P and the other elements (Table 6).
The repeatability was not checked by a separate experiment because the method requires duplicate samples to be run, and to compare within 5%. Besides, the overall 6day x 2
intermediate precision RSD was typically well below 5%. There are no SMPR criteria for intermediate precision, but these are certainly adequate and in agreement with the
measured reproducibility, which are a couple percent higher (Table 4).
The RSD(R) from the 2011.19 MLT performed in 2014. The RSDs for the 9 added elements are actually straight RSDs from the lab averages, but those should be pretty close to a
correct RSD(R). The Level column is filled in with "not relevant" because all the SPIFAN materials were pretty close in concentration of a given analyte on an RTF basis, and nothing
was at the low end. The only analytes we should probably look at "low" are Mn and Cu, and we did not do this by spiking a placebo. The RSD (R) all met SMPR criteria except for
Mg in one product and P in two products. Mg generally looks like it has no problem meeting a 10% criteria, but P will struggle to beat 8%. On the other hand, ICP-MS excels at the
copper and Mn RSD(R)s compared to ICP-AES.
Linearity criteria are not contained in the SMPR per se, but linearity certainly affects other criteria. Table 5 shows the 3-day averages of independent standard solutions run as
samples at various points along the curve. Typically, one would want to see 95-105% recovery from these standards (which allows expansion to 90-110% recovery in actual
samples). The fact that P marginally passes the linearity test may be a clue as to why it does not appear to be determined as robustly as the other eleven elements.
Table 5b contains the PLOQ data derived from the Linearity study. These are transposed to the Method Evaluation Form. The PLOQs meet the SMPR except for Cu, Mn, and Fe.
The MTE-02 method is not designed to go 10x lower on these elements, but could likely be redesigned to do so since it is based on ICP-MS. As it stands, the calibration is
optimized for the higher levels of trace elements seen in all our products.
MTE-02 MEF Form
FOR ERP USE ONLY
DO NOT DISTRIBUTE