John

Your data is a worrying. We use the Roche method and perform well on EQAS but I have just had a look at the values that we get for the scheme and can see that the lowest values are usually around 0.7mmol/L whereas the problems seem to be most exaggerated at less than 0.4mmol/L.

There is a strong argument for EQAS schemes to include extremely low values from time to time to test out this issue.

Regards

Graham

 

 

We should be most grateful for any comments that Roche users, or others, may have on two issues that have arisen in our laboratory.

 

1) HDL Cholesterol

In January we changed our main analytical platform from Olympus to Roche (cobas 6000), and initial evaluation data for the new HDL cholesterol method was acceptable (Roche = 1.04*Olympus - 0.08). However, we have now become aware of a small subset of patients with HDL values toward the lower end of the range whose results are significantly lower when measured by the Roche method. Comparison of the first 9 months of Roche data with the previous 9 months of Olympus results showed overall distributions of cholesterol and HDL that were virtually identical, but there were many more very low (<0.15 mmol/L) HDLs  in the Roche group  and the number of total:HDL cholesterol ratios greater than 10 was four times higher:

 

Median data (results in mmol/l):

	Lipids	Chol	Trig	LDL	HDL	Ratio	Ratio >10
Roche	30217	4.9	1.4	2.7	1.32	3.6	77
Olympus	35115	4.9	1.4	2.7	1.35	3.6	18

 

 

 

 

 

	Olympus	Roche
Lowest HDL	0.23	0.09
Highest Chol/HDL ratio	16.5	47.3

 

An example of serial results for an actual patient is shown below, where the last set was obtained using Roche reagents:

 

  Date\Time       CHO   TRIG    HDL    LDL  CHO/HDL

 

 180805 0835      4.8    2.8   0.42    3.1   11.4  Olympus

 230206 1230      4.9    3.7   0.32    2.9   15.3     “

 040507 1000      4.8    3.9   0.37    2.7   13.0     “

 231107 0920      3.8    3.0   0.42    2.0    9.0     “

 171108 0910      4.8   10.2   0.51     NA    9.4     “

 230909 0900      5.2    2.5   0.15    3.9   34.7  Roche

 

Also results for 2 samples comparing results obtained by Roche methods with those from a different analyser in a neighbouring laboratory:

 

   CHO   TRIG    HDL    LDL  CHO/HDL

 

 Sample 1 Roche      4.6    2.3   0.30    3.3   15.3 

 Sample 1 Other      4.6    2.6   0.64    2.8    7.2    

 

 Sample 2 Roche      3.5    3.9   0.11    1.6   31.8    

 Sample 2 Other      3.5    4.1   0.30    1.3   11.7    

 

Have any other Roche users noted this problem, and if so do you have an explanation or work around for it?

 

2) Magnesium

Earlier this year Roche circulated a Product Alert Notice for cobas systems requiring adjustment of C.f.a.s set point for magnesium. This re-standardisation was necessary due to deviation of the master-lot set point to atomic absorption spectrophotometry, the correction being in the order of minus 10%. Currently the mean of the Roche chlorophosphoazo III dye method group is close to the all method mean for magnesium in the UKNEQAS EQA program, although this scheme does not report any results for atomic absorption methods. A reduction of 10% would put the Roche group out of line with other scheme participants.

Have other Roche users independently verified the need for this change, and is it justified by the method’s performance in other (non UK?) EQA schemes?

 

Many thanks

 

John Sale

 

Dr J K Sale

Principal Clinical Scientist

Clinical Biochemistry Department

Poole Hospital NHS Foundation Trust

 

Tel:      +44 1202 448067

E-mail: [log in to unmask]

 

 

 

 

 

 

Dr J K Sale

Principal Clinical Scientist

Clinical Biochemistry Department

Poole Hospital NHS Foundation Trust

 

Tel:      01202 448067

E-mail: [log in to unmask]

 

 

 

 

------ACB discussion List Information-------- This is an open discussion list for the academic and clinical community working in clinical biochemistry. Please note, archived messages are public and can be viewed via the internet. Views expressed are those of the individual and they are responsible for all message content. ACB Web Site http://www.acb.org.uk List Archives http://www.jiscmail.ac.uk/lists/ACB-CLIN-CHEM-GEN.html List Instructions (How to leave etc.) http://www.jiscmail.ac.uk/