An interesting paper. Whether thyroid tests need to be done fasting - or not - has been discussed many times.
If we accept this paper, we can see that fasting versus high blood glucose seems to have a potentially very significant effect - with TSH falling and FT4 rising.
Ironically given that we are often told that FT3 varies too much to be useful, the impact on FT3 is relatively slight.
However, the authors come up with a totally nonsensical idea in their conclusions:
Different reference ranges can be used for thyroid values based on whether the patient is in a fasting or fed state.
I simply do not see how you can do that as people consume different foods, in different quantities, and at different times before having their blood drawn. You cannot reduce that to two ranges - you would need many. Effectively, I think you would have to measure blood glucose and use that to dictate which reference range to use or derive a "correction" factor. If you think we have problems now with different labs having different ranges, just imagine that each one had different ranges (or approaches to correction factors) for some or all the various blood glucose levels they try to consider.
Therefore, in my view, there is just one sensible approach. Always have your blood draw fasting and a significant time after your last intake of calories.
My opinion has long been that we should fast for thyroid tests. My rational is that some tests definitely have to be done fasting. INR, blood glucose and many others.
If we always fast then, should we need one of these tests which must be done fasting, then any impact of fasting versus not fasting, even if it is very small, is irrelevant. Because our entire history of blood tests will have been done fasting.
Effect of Pre- and Postprandial Plasma Glucose Levels on Thyroid Hormones: A Cross-Sectional Study
Abstract
Background
There is not much data regarding the effect of plasma glucose on thyroid hormones. Currently, there is no consensus regarding the timing of blood sample collection for thyroid hormones. Evaluation of the correlation between plasma glucose and thyroid hormones may enhance pathophysiological understanding of postprandial thyroid stimulating hormone (TSH) decline.
Objectives
To study the effect of pre-and postprandial plasma glucose levels on thyroid hormones.
Methodology
A cross-sectional study was done among participants aged 18 years and above after obtaining informed consent. Venous blood samples for fasting and postprandial plasma glucose, fasting, and postprandial thyroid profile ((FT3), (FT4), and TSH) were collected. The association was studied using the t-test and chi-square test between groups, while correlation using Pearson’s correlation coefficient. A p-value of <0.05 was considered statistically significant.
Results
Among the 197 participants, 126 (64%) were female and 71 (36%) were male. The mean (± S.D.) of age was 43.74 (± 12.62) years. Mean (± S.D.) postprandial TSH (4.31 μU/ml (± 7.79)) and free T3 (285.52 pg/dl (± 32.46)) were less than the fasting values (4.99 μU/ml (± 7.78)) and (295.84 pg/dl (± 32.56)). The mean (± S.D.) of both fasting and postprandial TSH and free T4 were less in the high plasma glucose group compared to the normal plasma glucose group (fasting state p-values 0.005, <0.0001 & postprandial state p-values 0.016, <0.0001). The correlation between fasting plasma glucose and fasting TSH values (Pearson correlation; r=-0.18; p-value 0.232) was observed across all the plasma glucose ranges.
Conclusion
There is a significant postprandial decline in TSH levels than the fasting TSH, indicating that there is a plasma glucose-mediated reduction in TSH values.
Cholesterol testing is supposed to be done fasting apparently. When was the last time any patient in the UK was told that either? American endocrinologist Lustig says the test is worth absolutely nothing if done not fasting.
AND even more importantly, those of us hypos, he flatly said, that has to be fixed (as do our NICE Guidelines, ignored by every doctor I have ever seen) before any cholesterol test will mean anything.
Unfortunately no idea what he means by that, since hypo testing is also ‘broken’.
Before 2014, fasting samples were used for a lipid profile, but since then fasting is no longer routinely required. However, fasting may still be needed in some circumstances, so you should follow your doctor's advice.
I guess all our bodies changed in 2014. (Without more explanation, we are left making ridiculous statements and not understanding the significance of the date. I can't imagine that the tests changed because that would imply that every single lab in the UK changed within that year. And no other tests ever change that fast.)
Effects of calorie intake and sampling time on thyroid stimulating hormone concentration
Aimei Dong, Youyuan Huang, Yucheng Huang & Bing Jia (Published: 01 April 2022 in BMC Endocrine Disorders)
"Discussion
The findings of this study showed that the TSH level was reduced significantly by about 30% after calorie intake in the morning, The components of calories had no significant influence on TSH variation rate when the calories intake was similar. The TSH level was reduced slightly by 5.2% in the subjects after maintaining the fasting state. The rate of TSH reduction was significantly pronounced after calorie intake compare to the fasting state, suggesting that the influence of food on TSH was more evident than the diurnal rhythm of TSH.
The findings of this study showed that the variation of TSH level after calorie intake in the morning might influence the diagnosis of subclinical thyroid dysfunction. Subjects with subclinical hypothyroidism might be underestimated due to the non-fasting state.
Conclusion
In summary, the TSH level was reduced significantly after food intake, compared with that at fasting state in the morning. If the reference range of TSH used in the laboratory was from fasting blood samples, it would be better to evaluate the TSH level in fasting blood obtained in the morning compared with random or postprandial samples."
Effectively, they all have to re-evaluate their range setting. Because, until they do, they can never really know anything.
They really should look at timing and fasting in all blood tests which have hitherto been regarded as time and/or fasting independent but have, so far, not been subject to rigorous proving trials.
need to read it more carefully, may have missed it .... but this one doesn't seem to control for TSH lowering due to time of day ? ....
if they test ,then eat , wait a bit , and then re-test ... it's later .....so TSH would be lower anyway. (i think the one i posted did have a control group for this )
But time of day is very difficult. The daily rhythms paper showed that some people have very little variation whereas others have quite substantial variations.
If time of day is used blindly, it would likely negatively affect those who exhibit little variation even as it positively affects those with a substantial variation. And we also need to consider the fundamental differences between those who are taking thyroid hormones and those who are not.
I'm very much not a fan of AI in general. But there are some areas where I think something between AI and hand-crafted code - or both together - might actually be able to handle all these factors - when ordinary humans can't. At least, can't in the context of having a few minutes to see the results and discuss with patient and prescribe and/or advise. And similarly for iron tests which are incredibly difficult to interpret.
yes there is a lot to juggle ...... i can't do it , because my breakfast was substandard ,,, i wanted grilled belly pork with apple sauce....... i could have gone out and got some ... but it's very chilly out there. .. so i made do with Hummous on toast , thus my mental faculties are diminished .
.. "alexa ~ go out and fetch me a nice fat bit of pig"
oh dear that hasn't helped ... now i can't think of anything but belly pork.. i may have to go out and get some and call it 'lunch'.
helvella 'If we always fast then, should we need one of these tests which must be done fasting, then any impact of fasting versus not fasting, even if it is very small, is irrelevant. Because our entire history of blood tests will have been done fasting.'
I completely agree. Especially given that we frequently see people reporting that, unbeknown to them, a thyroid test was sneaked in alongside some other test/s. The difficulty though, can be getting booked in for a test at a time of day that works for your own 'fasting' regime. For most of us, that would conveniently be early morning, having not eaten since the evening before. But it's not always easy to get that early morning appointment on an appropriate day.
One traditional problem was all the people on Warfarin needing their fasting INR tests. Because that was recognised, they were given priority over everyone else. I think fewer of these are done these days. But still a problem.
I think we could imagine a future in which we get issued with devices like the Tasso. (Mentioned here many times previously.) So we could - effectively - ALL get fasting blood draws at a convenient time.
If the cost of Tasso-like devices all round drops below that of traditional phlebotomy, it does seem feasible. Leaving human phlebotomy appointments to those who need them - such as needing more blood, unable to cope, etc.
i saw a piece on telly the other day .. one of the london? hospitals is using a fleet of drones to fly urgent blood samples to the lab (for thing like transplants etc ) , idea came from a doctor from Afghanistan who grew up accustomed to drones dropping other things... thought he'd turn it round to do something useful.
So we could have a Tasso and book a drone to pick it up ....
Content on HealthUnlocked does not replace the relationship between you and doctors or other healthcare professionals nor the advice you receive from them.
Never delay seeking advice or dialling emergency services because of something that you have read on HealthUnlocked.
Effect of low calorie diets on thyroid function
