Sensitive, specific test for insulin resistance

Investment Analysis Clubs Macro Economic Trends and Risks
1

The Triglyceride glucose (TyG) index, a sensitive, specific test for insulin resistance has Macroeconomic impact because it is easy to compute and has a very high correlation with many expensive chronic health conditions.

This has just been published in the European Journal of Internal Medicine, 2025, so it’s very current.

https://www.ejinme.com/article/S0953-6205(24)00375-3/fulltext

image
image1290×927 67.6 KB

It’s very easy to calculate from a simple blood test.
The calculation for your TyG index used the following equation:

TyG Index=ln(Fasting Triglycerides (mg/dL)×Fasting Blood Glucose (mg/dL)/2)​

Where:

  • ln represents the natural logarithm.
  • Fasting Triglycerides are measured in milligrams per deciliter (mg/dL).
  • Fasting Blood Glucose is measured in milligrams per deciliter (mg/dL).

A high TyG index correlates closely with the speed of cognitive decline in Alzheimer’s disease. An increase in TyG index has also been correlated with amplified risks of all-cause mortality and cardiovascular death among individuals at high CVD risk.

This is one number that every person should know. It’s very easy to manipulate triglycerides with lifestyle. I reduced my triglycerides from 175 to 95 by simply giving up drinking ginger ale. No meds required.

Here’s a table of TyG levels.

|TyG Index Level|Interpretation|
|Below 4.5|Generally considered within a healthy range, indicating lower insulin resistance.|
|4.5 or higher|Often associated with increased insulin resistance and higher risk of metabolic conditions.|
|> 8.7 or > 9.04|Higher values (e.g., above 8.7 or 9.04 in some studies) are often strongly linked to increased risk of type 2 diabetes, metabolic syndrome, and cardiovascular disease.|

The amount of suffering (and cost) that could be averted by lifestyle changes is staggering.
Wendy

5 Likes
2

My TyG in that case is - ln(71 x 110 / 2) = ln(3905) = 8.27

I don’t see how it could possibly be 4.5 in any case. Because e^4.5 is only 90, and triglycerides multiplied by glucose and then divided by 2 can’t possibly be only 90 even with the best of numbers for those two items.

I asked an AI and it came up with the same conclusion that I did. Under 4.5 is not realistic for a typical living human being. I’m guessing that we aren’t understanding the units (or the calculation, or the ranges) properly.

Here is what the AI said at the end of the whole conversation -

Conclusion:

It’s difficult to find typical numbers for triglycerides and glucose that result in a TyG below 4.6, as this would require exceptionally low values (e.g., triglycerides <10 mg/dL and glucose <20 mg/dL), which are not typical or medically realistic. If you meant a different target (e.g., <8.6), please clarify, and I can provide more typical examples! For reference, your earlier values (triglycerides = 71, glucose = 110) gave a TyG of 8.27, which is within a common range.

3 Likes
3

It’s TyG = ln [Fasting triglyceride (mg / dl) x Fasting glucose (mg / dl)] / 2
Therefore ln(71 x 110) / 2 = ln(7810)/2 = 4.48

There was a typo in the equation in Wendy’s figure above.

Mine’s 4.43. Triglycerides 71, Glucose 99. My glucose has been hovering around 100 for 15 years.

5 Likes
4

Now this is particularly hilarious. And it illustrates a HUGE problem with AI in general. That’s because I didn’t tell the AI (both grok and perplexity) how to calculate TyG (and I didn’t point them to the paper or anything at all), I just asked them to figure it out for me. That typo is out there in the world, it’s NOT Wendy’s typo, it’s a typo that is present all over the place. And the AI learned using that typo!!! Now, in the last 5 minutes I told all the AIs that it is a typo and explained to them how to do the calculation properly. I hope that when others use the AI in the future, it will use the correct version without the typo. If I remember, I’ll check again in a few days.

So my TyG is 4.48, no quick Alzheimer’s for me! :rofl:

5 Likes
5

I didn’t say it was Wendy’s typo, just a typo in the figure.

4 Likes
6

There are online calculators for this index so I used one and the last trig and fasting glucose without lipid lowering meds a few years ago…78 and 95. Calculator gave me a TyG of 4.46.

I knew I was pretty insulin sensitive … as I ought to be given my lifestyle choices (assuming that to be the only factor :wink: … as if) as, at that same appointment, I asked for a fasted insulin measurement too since just a fasted glucose and even an A1c doesn’t tell you whether that “healthy” glucose is achieved with high circulating insulin…as would be the case with any degree of insulin resistance…or a healthy insulin response and decent sensitivity. I think my husband calculated my HOMA IR at .9.

Edit…for the heck of it I checked my most recent trig and glucose values now my lipids are way down and it was 4.23. No reason to conclude that I’m any more insulin sensitive…just that the variables have been manipulated a bit

7

That’s great info, Wendy. Insulin resistance could be causing many chronic diseases and lifestyle changes to prevent it may help many people.

8

Thank you for catching that. I was wondering why mine kept coming up at 7 something even though my triglycerides were well within normal limits.

9

@Volucris thank you for pointing this out. It’s reasonable.

But I copied and pasted the equation directly from the original article so if there was a typo that is the source. It was a review article so that’s possible.
Wendy

10

@MarkR where did you find the “corrected” equation? Do you have a link? I copied and pasted directly from the review article that I linked.

If you want to “correct AI” you will have to include Google Gemini and ChatGPT.
Wendy

11

@Volucris my numbers are almost identical to yours. Based on the original equation from the article my TyG is 8.25 which shows a concerning level. I wasn’t surprised by that since my A1C shows that I’m pre-diabetic even though my fasting glucose is 95. My BMI is 26 and I have some central fat.

My take-away wasn’t that the equation was wrong but that I need to lose some weight and bring my A1C down.

But if the equation has a typo I don’t have anything to be concerned about.
Here is the original source of the Triglyceride-glucose index (TyG index) .

The product of fasting glucose and triglycerides as surrogate for identifying insulin resistance in apparently healthy subjects

by Luis E Simental-Mendía 1 , Martha Rodríguez-Morán, Fernando Guerrero-Romero,
Metab Syndr Relat Disord

. 2008 Dec;6(4):299-304.
doi: 10.1089/met.2008.0034.
Metab Syndr Relat Disord

Background: Because the insulin test is expensive and is not available in most laboratories in the cities of undeveloped countries, we tested whether the product of fasting triglycerides and glucose levels (TyG) is a surrogate for estimating insulin resistance compared with the homeostasis model assessment of insulin resistance (HOMA-IR) index…

The best TyG index for diagnosis of insulin resistance was Ln 4.65, which showed the highest sensitivity (84.0%) and specificity (45.0%) values. The positive and negative predictive values were 81.1% and 84.8%, and the probability of disease, given a positive test, was 60.5%…

[end quote]

This is the original source article. Unfortunately the entire article is behind a paywall so I couldn’t see the specific equation.

Here’s another study.

Association between the triglyceride glucose index and in-hospital and 1-year mortality in patients with chronic kidney disease and coronary artery disease in the intensive care unit

Cardiovasc Diabetol, . 2023 May 13;22(1):110.
doi: 10.1186/s12933-023-01843-2.

Objective: This study aimed to explore the association between the triglyceride glucose index (TyG) and the risk of in-hospital and one-year mortality in patients with chronic kidney disease (CKD) and cardiovascular disease (CAD) admitted to the intensive care unit (ICU).

Methods: The data for the study were taken from the Medical Information Mart for Intensive Care-IV database which contained over 50,000 ICU admissions from 2008 to 2019.

The TyG index was computed according to the triglyceride (TG) and fasting blood glucose (FBG) concentrations according to the equation: ln [TG (mg/dl) × FBG (mg/dl)/2].
[end quote]
The average TyG of these patients was about 9. One out of six died over the course of a year. The response was non-linear. The participants were divided into four groups based on various TyG indices: T1 (TyG ≤ 8.716), T2 (8.716 < TyG ≤ 9.037), T3 (9.037 < TyG ≤ 9.664), and T4 (TyG > 9.664). The Kaplan–Meier analysis plot showed a significant difference among various TyG index groups. The TyG index group T4 had double the risk of death of the others. They were very sick to start with (in the ICU with both Coronary Artery Disease and Chronic Kidney Disease) and they died very quickly.

image
image1170×1032 89.4 KB

In conclusion, it seems like the original research does confirm the equation that shows that both of us have somewhat high TyG values which indicate some insulin resistance. However, the practical impact of these borderline high values is pretty small. The real deadly effect shows up in people who have very high TyG values.

Wendy

1 Like
12

Actually, I think that even without the equation, there’s stuff to be concerned about in your own metabolic profile…a high BMI with belly fat and a pre diabetic A1c. Requesting a circulating insulin test at your next doctor’s visit might not be a bad idea.

I hung out on the old Diabetic Fools board for years while I still had my practice and since. It was a great source on insight into management of both Types 1 and 2 diabetes…and the progression to the latter from “barely” pre diabetic.

I think it was Fuskie who would often say that “pre-diabetic” was something of an artificial construct somewhere on the path from a healthy metabolic response to outright disease. The thing to be aware of is that it’s not just the high circulating glucose that causes the damage to the endothelial lining of your blood vessels but the excessive levels of circulating insulin that the pancreas is pumping out in an effort to overcome increasing insulin resistance too.

4 Likes
13

Here you go, a pretty comprehensive article on insulin and IR with some historical perspective to boot.

Trends in insulin resistance: insights into mechanisms and therapeutic strategy | Signal Transduction and Targeted Therapy Trends in insulin resistance: insights into mechanisms and therapeutic strategy | Signal Transduction and Targeted Therapy

There’s plenty of readable insight here such that the molecular stuff shouldn’t be too much of a turn-off even for the uninterested (although I guess they probably won’t bother reading it anyway)

2 Likes
14

So, just out of interest and because I’m a fair way from being as mobile as I’d like, I squandered some googling time on trying to get a grasp on the merits of this biomarker as a tool (if there are any…I’m wary of concluding that on the say so of study authors…I think “Mandy Rice-Davies Applies” here). Specifically to see if there’s much available comparing this index with what’s already available to assess potential for early metabolic dysfunction…including, but not limited to, bathroom scale, mirror, and tape measure.

Seems there is…it’s not a new biomarker…quite a few studies comparing it with HOMA-IR, A1c. Here’s the odd thing, these studies have pretty much all been from SE Asia…China, Korea etc…like this in the OP. So, I got to wondering…like you do…if this is a biomarker that’s a superior measurement in that population (given the differing predisposition towards metabolic disease across different demographics) Then I lost interest.

15

The article notes that TyG is basically a free test for anyone who has already had a common glucose test and lipid panel since it only requires a calculation. It appeals especially to doctors in countries that don’t have the resources to order a circulating insulin test.

Wendy

16

Oh I can see that as an excellent reason to use it in those countries. Especially if it additionally gives insight into how that particular demographic might find greater value in its use…i.e. being more predictive than, say, A1c or any other biomarkers that folk have access to here.

17

Thank you for this valuable information, @VeeEnn. My younger sister will also be interested. Like you, she may have a genetically-determined condition since she’s Size 6, follows a keto diet, works out every day and still has pre-diabetic A1C. Our father was diabetic.
Wendy

2 Likes
18

This hyperinsulinemia is a very early feature in insulin resistance, apparently and is actually why I requested the circulating insulin test at the time I mentioned the CAC scan possibility to my new PCP. I also added that I would pay our of pocket should Medicare not pay. My A1c was routinely around 5.4 or so and not rising, so I could see that. I was just trying to cover every single possibility for why my LDL-C was raised (enough to bug me, obviously, if not my PCPs)… and even with my CAC scan, I was hoping to put my mind at rest.

Well, we see how that turned out!

19

Here are a couple of the podcasts that opened my eyes to the importance of insulin resistance…and how much I didn’t know about the topic. I think there’s enough on both of them to highlight the complexity of the condition and its role in metabolic dysfunction before the paywall kicks in.

#140 - Gerald Shulman, M.D., Ph.D.:
A masterclass on insulin resistance—molecular mechanisms and clinical implications - Peter Attia #140 - Gerald Shulman, M.D., Ph.D.: A masterclass on insulin resistance—molecular mechanisms and clinical implications - Peter Attia

#337 - Insulin resistance masterclass: The full body impact of metabolic dysfunction and prevention, diagnosis, and treatment | Ralph DeFronzo, M.D. - Peter Attia #337 - Insulin resistance masterclass: The full body impact of metabolic dysfunction and prevention, diagnosis, and treatment | Ralph DeFronzo, M.D. - Peter Attia

The Gerald Shulman one was interesting as he emphasized how early on it can start. He’s at Yale and quite early in his research career, he recruited a bunch of undergrads there…all lads in their late teens, early twenties. None had the Usual Suspects in the context of anything amiss on their standard bloodwork, body composition etc. None had any close relatives…parents, siblings…with any history of T2D. Jocks were also excluded. So basically a group of regular sedentary, students, probably eating the same sort of craptaculous diet…but who were probably at the point in their lives when they could get away with it. Or so they thought. I think the figure was something like 80% of them already showing early insulin resistance.

IIRC, these topics are pretty well covered in his Outlive book

2 Likes
20

The correct formula:

The other formula was giving unreasonable answers.

2 Likes