Post Number: 253
|Posted on Sunday, February 10, 2008 - 11:04 pm: |
The results of this study support the use of the Lactate Pro as an
accurate and reliable method of measuring blood lactate under a
variety of environmental conditions. This supports the previous
work of Pyne et al. . In their work they investigated the Lactate
Pro in a laboratory and field setting against a variety of other
analysers as well as against another Lactate Pro analyser. The correlations
between the two analysers in their work, when compared
against the current work are the same (r = 0.99). The correlational
data between the two Lactate Pro analysers would suggest
that these analysers can be used interchangeably within the
same testing sequence, although, of course, it is recommended
that the same analyser is used with the same subjects throughout
the testing condition.
Makita and Satomi  have found the machine to have a coefficient
of variation of two percent for low concentrations and 4.7
percent for high lactate concentrations. We however found much
larger CV figures for the low concentrations based upon 20
samples and similar values for the moderate and high lactate concentrations
(4.1 and 3.1% respectively). The reported CV for the
instrument by the manufacturer  is 3%, which approximates
our figures, especially within the 2–10mM range, which is the
training and testing range. We would conclude that these higher
CV values are not particularly important for sports related testing,
as use of this analyser would be predominantly in a field testing
situation and lactate values would typically be in the lower regions.
The reliability measurements are similar to those reported
for the Accusport, 7.0% and 4.6% for similar values. These reliabilityvalues
areconsidered acceptable foranalysers that are portable
and for use in predominantly field settings by untrained staff .
It was interesting to note that when a comparison of the
pooled Lactate Pro data (from 6 and 8) was made with the Kodak,
as per the work of Fell et al. , the regression line was
y = –1.009 + 0.887 N X, which would give significantly different
values to that obtained in the Fell et al. study .
When a comparison of the machineswas made, Lactate Pro number
8 only correlated highly with the Accusport (r = 0.834) in the
humid condition, which is interesting since this was the only
other portable analyser compared. This contradicts the work of
Makita and Satomi . In their work they found a high correlation
(r = 0.99) whereas we found a much lower correlation
(r = 0.785). Makita and Satomi’s  work is also supported by
Pyne and colleagues  who also found a correlation much
higher than ours (r = 0.973). It may be possible that improvements
in the data for the Accusport could be made by pipetting
an exact amount of blood to the test strip rather than simply
using a “drop” of blood.
The analysis of variance (ANOVA) that was undertaken on the
pooled data from the machines across the three conditions indicated
that the machines measured significantly differently, with
only the Lactate Pro and Accusport not being significantly different.
This is interesting given the low correlation between these
two analysers. However, when a square root of the error component
of the MANOVA conducted on the datawas calculated, a value
of 1.1 mmol/L was seen. This suggests that the random error
from analyser to analyser is 1.1 mmol/L. The systematic bias,
however, is ruled out due to the non-significant MANOVA [3,10].
The difference in random error of 1.1 mmol/L may be due to errors
in blood volume seen with Accusport analyser, as we did not pipette
the volume to the analyser strip, whereas the Lactate Pro
aspirates 5 mL automatically, therefore eliminating volume error.
Significantly, the Analox measured lower and the Kodak higher
than the Lactate Pro with the Analox measuring lower than the
Kodak and Accusport. This data suggests that these machines,
apart from the Lactate Pro and the Accusport, do not give the
same result, despite being moderately or highly correlated.
An ANOVA carried out in the normal condition across all workloads
indicated that there were significant differences suggesting
that the machines measured lactate concentrations differently
across the tests. The post hoc analyses carried out indicated that
the Lactate Pro analyser differed significantly from the Analox,
with the Analox also measuring significantly lower than the
Kodak. Thiswas also true for the Accusport, which also measured
significantly lower than the Kodak. The samewas not true for the
analysers in the hot condition where there were no differences
between the machines. Why this might be so is a reason for conjecture
and as yet we are unable to quantitatively explain this
difference. In the humid condition, although the ANOVAwas significant,
it was on the basis of a significant difference only between
the Analox and Kodak machines.
The four machines, as shown in Fig. 2, cannot be used interchangeably
across different workloads as they would give different
results, since the slopes of the lines as shown are all different.
Although speculative, we suggest this may be due to the method
of measurement used by each machine. Certainly, the Kodak Ektachem
uses plasma and it is well known that plasma values differ
significantly to whole blood [17, 20].
The use of correlational techniques has been questioned because
they are measures of relation rather than agreement [1,3,13] and
as such can be highly influenced by the range of subject measurement.
Bland and Altman  have proposed a different method,
which we have used, termed “limits of agreement” . Our
data shows that there are few of the data with a “perfect agreement”
(equal to 1.00). In Table 3 (Exp SD LD) there is a great deal
of variation between the data suggesting that few, if any, of the
machines should in reality be compared to each other. In all of
the three conditions, the two Lactate Pro analysers have the least
amount of variation (17%, 13% and 12% in the normal, hot and
humid conditions respectively). The worst combination to have
would be the Analox/Kodak (normal and humid) and the Analox/
Accusport in the hot condition .
In conclusion the Lactate Pro analyser has been shown to be a
simple and effective measurement device for taking blood lactate
in a field setting. Depending upon the method of statistical
comparison, it compares favourably to other similar devices, as
well as to those found on the laboratory bench. We would caution
against using this machine, or for that matter combinations
of other machines, to compare data until the individual laboratory
has assessed the relationship between the two machines
used for measurement under the conditions found within the
laboratory or field situation. In real terms it would be best to
use the same machine under all circumstances.
Post Number: 283
|Posted on Sunday, February 17, 2008 - 03:18 pm: |
Thanks for that question : Here the summary
How do we know , that the Lactate pro is accurate in repeating blood sample, when it takes 1 min till you have the result. This is exactly the question we have to ask.
Here some very simple tests and you see why in the Analox study the blood levels with the equipment possibly was always lower than in the Lactate pro or in similar equipment.
1. We test 2 / 3 or more lactate pro at the same time. . Same time meaning same time in the possibly same second.
This is not or very difficult to do with the analox but could be done close.( Handling problem ).
We poke 3 fingers at the same hand at the same time and take 3 lactate pro and take the blood more or less in the same second.
The 3 lactate pro's will give very much the same result . I had demos where I was o lucky that I had 3 equal readings but normally it is o.1 o.2 of.
2. You take resting lactate by the same person with every time a new finger poke and you will have equal plus minus o.1 - o.2 results.
Try that with an anlox or YSI with glass tubes and injection with 2 or 3 different lab technician. Good luck for al of them.
Now here the last experiment.
You take lactate after a hard 4 min run on a client. Take with the 3 lactate pro you tested and you had equal results.
You take one every 20 second so 40 seconds to 60 seconds difference from the initial poke to the last sample . You wil see depending on the clients lactate dynamic perhaps a picture like that.
6.3 6.1 5.6. and if you have 1 min difference perhaps up to 1 mmol and more difference from the initial blood test to the last sample all taken at the same finger no new poking just what is coming from there.
1 min difference can be 1 mmol or more different readings.
Now check the handling procedure with a YSI and or an analox. If you are very good you may be able to have the sample in your equipment in 1 min , if you are not that good and you take a sample and you have problem with enough blood in your glass capillary you have longer.
The difference will be as shown above.
That does not mean the analox is not a very accurate apparatus , in fact it may be as accurate as it can be, but how accurate is the human being doing the testing. The key is to avoid as much as possible human handlings.
Believe me , that a Diabetic one kid ( no certified lab or nurse ) just a daily practice of 10 and more tests will be much more accurate interesting his blood sugar than anything close what a nurse or lab technician can offer.
I have a son with diabetic I and he can take his blood with closed eyes in bed from his ear in a blink of an eye with not loosing a drop , no squeezing just perfect, and he get's nuts , when he is getting tested in a lab where the lab technician has to squeeze and nothing comes and finally has a bloody mess allover and the result should be more accurate.
After literarly over 20 thousand blood lactate samples i can say that the lactate pro allows us to make the smallest mistake possible with handling , but even still up to this days, if we have a strange readings it is me not the equipment , which failed.
Too much squeezing ,
not taking the first drop of the finger ( dirt on the skin )
too much alcohol for pseudo sterilization
Cold finger with pooled blood.
Pooled blood in a sport like biking or cross country skiing because of a very tight hand grip.
Too small drop and taking the analyzer too fast of the drop . wait till peep.
. Just to name a few simple handling mistakes, , which always can happen, and certainly if testing outside in the field.
So testing and comparing equipment should be done in a fair scientific way like the article I showed. What is the golden standard . Well not always the equipment which is longer on the market.
Remember the old swiss watch with the hundreds of cog wheels , very accurate but not as much anymore compared with high tech watches.
Now which one is the golden standard. ?
Okay "every body has a watch , but nobody has time."
This is a word from Julius Kariuki( Kenia ) the 3000 m steeple chase gold medal winner in Seoul. It is a story on it's own on the people in Switzerland with all their watches but no time for a live.