I'm on Growth Hormone and Afraid of Insulin Resistance. I Was Tracking the Wrong Blood Marker.

A guy on Reddit was thinking about starting growth hormone. Early fifties, on TRT, lean, wanted it for recovery and body composition and better sleep, and he asked how to go about it. I gave him a real answer, because I am on it too. Part of that answer was a warning: growth hormone pushes your blood sugar up and hurts your insulin sensitivity, so watch your fasting glucose. I told him to check it a couple times a week. I had been doing exactly that to myself.

Then it nagged at me the rest of the day. Is that actually the right test? I had picked that up years ago off a message board, it made complete sense to me, and I had never once questioned it. So I went back and did the research on my own confident answer. It was wrong enough that I went and edited my comment. I had been watching the right system at the wrong stage.

What growth hormone actually does to your blood sugar

The warning I gave him is real, and it is the reason I track this at all. Growth hormone is a counter-regulatory hormone, which is a polite way of saying it works against insulin. It pushes your body to break down stored fat and release it into your bloodstream as free fatty acids. That process is called lipolysis, and the rise in those fatty acids is what drives the insulin resistance. Your muscles and your liver stop listening to insulin as well as they used to. GH can nudge you toward the exact thing I am trying to avoid, so I wanted a number that would warn me early, while I still had room to back off.

The test I picked to be that early warning was fasting glucose. Sugar in the blood. If it climbs, I am getting resistant, and I drop my dose. Clean and simple, and it made complete sense to me.

The problem is that it is late.

Fasting glucose stays calm while your body works overtime to keep it calm

Insulin resistance does not start with high blood sugar. It starts one step upstream, and your body spends years hiding it from exactly the test I was relying on.

Here is the actual sequence. Your cells slowly stop responding to insulin. To force sugar into them anyway, your pancreas pumps out more insulin. That extra insulin works. It drags your glucose back down to normal. So you keep testing your fasting glucose, and it keeps coming back fine, while behind the scenes your pancreas is straining harder every year to produce that “fine” number. Your glucose only rises at the very end, once your pancreas can no longer keep up.

This is not a theory I came up with. A large study followed thousands of people for years before some of them were diagnosed with type 2 diabetes, and mapped exactly when each marker moved. Their insulin sensitivity had been sliding for about five years. Their pancreas had been ramping up insulin output to cover for it. And their fasting glucose stayed nearly flat until it made a steep climb only in the last three years or so before diagnosis. The sugar number was the last one to break.

So a normal fasting glucose does not mean you are insulin sensitive. It can mean your pancreas is doing overtime to make you look insulin sensitive. That is the part I had backwards.

One honest note, because you will see this oversold online. That head start is a few years, not the decade or two some people claim. And it is the typical pathway, not a guarantee. Some people lose the buffer faster than others. So an early-warning number buys you time, it does not buy you a permanent all clear.

The number that actually moves first: fasting insulin

If the early problem is your pancreas quietly making more insulin to hide rising resistance, then the early signal is that extra insulin itself. Fasting insulin goes up while your glucose is still sitting there looking perfect.

You can do better than reading insulin alone. There is a simple calculation called HOMA-IR that combines your fasting insulin and your fasting glucose into one estimate of how insulin resistant you are:

HOMA-IR = (fasting insulin x fasting glucose) / 405, using insulin in µIU/mL and glucose in mg/dL.

It is not a lab you order. It is math you do on two numbers you already have, and it tracks insulin resistance better than either one by itself. When I run my own panel through it, a fasting insulin of 5.8 and a fasting glucose of 84 give me a HOMA-IR of about 1.2. The thresholds people use for insulin resistance usually sit around 2 to 3, so as of that draw I am in a good spot. But the number I actually care about is not the 1.2 itself, it is which direction it moves on my next draw, on the same lab. For my growth hormone question specifically, this is the number most likely to flinch first, because GH-driven resistance shows up as your body making more insulin long before your glucose gives it away.

Where HbA1c fits, and why it is not the early warning either

HbA1c is the third number on my panel, and it is the one people trust the most, so it is worth being clear about what it does. HbA1c is roughly a three month average of your blood glucose. It is weighted toward the most recent month, and it is useful, but notice what it is an average of. Glucose. It sits even further downstream than a single glucose reading, so it lags even further behind. It only moves once your glucose has already been elevated for a while.

It also gets thrown off by anything that messes with your red blood cells, since that is what the test measures. Anemia and low iron can skew it in either direction and have nothing to do with your actual blood sugar. HbA1c is good for confirming that a problem has been going on for months. It is not built to catch one early.

So which one should you actually trust?

This is the part I did not want to hear, because I went in looking for a clean answer. Stop watching glucose, watch insulin instead. That is not the real answer, and pretending it is would be me making the same mistake in the other direction.

These three are not the same test three ways. They are three different stages of the same process:

• Fasting insulin and HOMA-IR are the earliest signal, the compensation phase, while everything still looks normal on a standard panel. For someone on GH, this is the one most likely to move first.
• Fasting glucose is the point where the compensation is finally failing. It is also the number that is actually doing damage. High glucose is what glycates and harms your tissues over time. It is not just a late alarm, it is the thing the alarm is warning you about.
• HbA1c confirms it has been elevated a while, and it is the number that decades of complication research are actually built on.

So you do not pick one. You read them as a timeline. HOMA-IR is the early-warning light. Glucose and HbA1c are the “this is real now” lights. My mistake was staring at the last light on the dashboard and ignoring the first one.

But what if my glucose does start rising?

Fair question. It is the one that started all of this, and there are two honest halves to the answer.

First, a single high fasting glucose is not an emergency, because that number is noisy. It moves with your last meal, your sleep, your stress, and if you are on growth hormone, with how long ago you took your last dose. I learned that one firsthand. I check my fasting glucose first thing in the morning, then take my GH. Back when I dosed at night, that morning reading would land between 110 and 120. When I switched to dosing in the morning, so my fasting glucose was being measured a full day after the previous shot instead of the morning after a nighttime one, it dropped to between 90 and 100. Same dose, same me, same glucometer at the same time of day. The only thing that changed was how many hours sat between my last GH dose and the reading.

That is the acute side of growth hormone. In the hours after you take it, GH raises your blood sugar, mostly through that same lipolysis route, pushing fatty acids into your blood and making you temporarily less responsive to insulin. Dose at night and your morning reading lands while that is still going on, stacked on top of the growth hormone your body naturally releases in deep sleep. Dose in the morning, after you have already measured, and the next day’s reading sits about a day out, once the effect has mostly cleared.

So a glucose that jumps is not automatically a sign you are becoming diabetic. It can be the drug doing its short-term, expected thing at the moment you happened to measure. What would actually worry me is a sustained rise, a fasting glucose that keeps drifting up across many mornings measured the same way, with HbA1c creeping and HOMA-IR climbing alongside it. That is the chronic problem setting in, not a single 115 the morning after a night dose. And whether it changes my dose is a conversation with my prescriber, not a call I would make off one number, or off a blog.

On the lipolysis question, because it is a good one: yes and no. When GH raises your glucose it is largely working through that lipolysis and fatty-acid pathway, so a GH-driven bump is a fingerprint of that pathway being active. But glucose is a downstream, non-specific number, not a lipolysis meter. The same lipolysis nudging your glucose up is also why GH helps you mobilize fat in the first place, so a rise is not a clean measure of how lipolytic you are becoming. It is the cost side of that pathway showing up. Read the trend, not the spike.

Why I am still not going to oversell fasting insulin

Being honest in both directions means saying where the early-warning number is weak, because it has real problems.

The fasting insulin test is not standardized between labs. My “9” at one lab is not your “9” at another, so the number only means something as a trend on the same lab over time, not as an absolute you can compare to a friend. There is also no official cutoff. The “your fasting insulin should be under 8 or 10” figures you see come from functional medicine circles, not from the bodies that write the diagnostic guidelines. And no one has proven that chasing an isolated fasting insulin number changes how you actually end up. It is an early signal, not a diagnosis.

Glucose and HbA1c earn their trust here. They are standardized, they are reproducible, they are what an actual diagnosis is made from, and they are tied to the outcomes that matter. Anyone selling fasting insulin as the one test that matters is overselling it, the same way I oversold glucose on Reddit.

What this actually changed for me

Here is the dumb part. I had been checking my fasting glucose a couple times a week, and my full panel already had all three anyway. Fasting glucose, fasting insulin, HbA1c, every draw. I was fixated on the one number that moves last and ignoring the two sitting right next to it.

So I went back and edited that Reddit comment. I told him what I just told you. Fasting glucose is the late one, fasting insulin and HbA1c catch it sooner, and I was changing what I run myself.

Now I calculate my HOMA-IR from the insulin and glucose I already get, and I watch that trend over time on the same lab. That is my early-warning line for the growth hormone. If HOMA-IR starts drifting up, that is my cue to pay attention while my glucose is still fine. If my glucose or HbA1c actually start climbing, that is not an early warning anymore, that is the problem already at the door, and that is a real conversation with my doctor about the dose.

I did not have to order a single new test. I just had to read the labs I was already getting, in the right order.

If you are on anything that pushes insulin resistance, whether that is growth hormone, some other compound, or just the normal way this creeps up on people, get all three on your panel and read them as a sequence instead of one hero number. And take the actual decisions to an actual doctor.

I am not one. I am a guy who gave a confident, incomplete answer on the internet, and the only reason I know any of this is that I went and checked my own bad answer. Do that more than I used to.

Frequently asked questions

Sources

• Tabák 2009. The Whitehall II trajectory study. Insulin sensitivity had been declining for about five years before a type 2 diagnosis, with compensatory insulin secretion covering for it, while fasting glucose stayed nearly flat until a steep rise in roughly the last three years. Lancet 2009.
• Matthews 1985. The original homeostasis model assessment paper, the source of the HOMA-IR formula used here. Diabetologia 1985.
• Weir & Bonner-Weir 2004. The five-stage model of evolving beta-cell dysfunction: compensation first, decompensation later. Diabetes 2004.
• Nathan 2008. The ADAG study translating HbA1c into estimated average glucose, the basis for reading HbA1c as roughly a two to three month glucose average. Diabetes Care 2008.
• Growth hormone and insulin resistance. Review evidence that GH is counter-regulatory and drives insulin resistance largely through lipolysis and free fatty acids. J Clin Endocrinol Metab 2009.
• Acromegaly HOMA-IR meta-analysis. Insulin resistance measured by HOMA-IR as an early feature of growth hormone excess. PMID 33085039.

This is my own experience and general education, not medical advice. It is not a recommendation to start, change, or stop growth hormone or any other medication, or to act on any lab value. My prescriber is aware of my bloodwork and of what I am doing. Every person has different physiology, labs, and risks, so talk to a qualified medical provider about your own.