GKI testing for therapeutic ketosis and cancer treatment (1 of 1)
Home Metabolic Dietary Therapy: Fasting, Keto and Functional FoodsFasting The Complete Guide to Glucose-Ketone Index (GKI) for Therapeutic Ketosis

The Complete Guide to Glucose-Ketone Index (GKI) for Therapeutic Ketosis

by Maggie Jones

I talk a lot about my GKI on this blog. I’ve been tracking it for about 10 months now and it’s taught me so much about the way my body metabolizes different foods and the way my metabolism reacts to certain events like stress, illness, and exercise. I credit most of my diet success to tracking as it’s helped me discover individual sensitivities like protein and processed food. Best/ worst of all, there’s no way to cheat a blood test!

I cover a lot in this article so feel free to skip to what’s relevant to you:

What is the Glucose-Ketone Index (GKI)?

The Glucose-Ketone Index was originally conceived by Dr. Thomas Seyfried and his team at the University of Boston to track overall metabolic health in cancer patients. It was quickly adopted as a benchmark for those with other metabolic conditions and anyone who wants to keep tabs on their metabolic health.

Your GKI is the ratio of your blood sugar to blood ketones at any given point in time.

Tracking blood sugar or blood ketones separately can also be useful but the ratio provides added insight and a quick benchmark for someone pursuing therapeutic ketosis, like me.

The results showed a clear relationship between the GKI and therapeutic efficacy using ketogenic diets and calorie restriction.  The GKI is a simple tool that can help monitor the efficacy of metabolic therapy in preclinical animal models and in clinical trials for malignant brain cancer and possibly other cancers that express aerobic fermentation.

Meidenbauer, Joshua & Mukherjee, Purna. (2015). The glucose ketone index calculator: A simple tool to monitor therapeutic efficacy for metabolic management of brain cancer. Nutrition & metabolism. 12. 12. 10.1186/s12986-015-0009-2.

How to calculate your Glucose-Ketone Index (GKI)

GKI is blood glucose (mmol/L) divided by blood ketones [BHB] (mmol/L). If you’re in the US, it is blood glucose (mg/dL) divided by 18 divided by blood ketones [BHB] (mmol/L).

First, you’ll need a blood glucose meter and a blood ketone meter. Blood glucose meters are pretty easy to find at any drug store and their test strips are usually inexpensive, as low as US$0.30 per strip. You may need to shop online for a ketone meter and those test strips usually run around US$2.00 each. I’ve been very happy with my keto mojo combined meter – just one pouch to carry around and the strips are around US$0.50 and US$1.00, respectively, on Amazon.

Getting a reading requires a little finger prick with the provided lancet and a 5-10 second wait for the results. Plug those numbers into the formula and voila, you know your GKI.

[ Get 15% off a Keto Mojo Meter Here. ]

After using it for about a year, however, I realized I could save money, keep track of fewer things, avoid the finger pricks, and get more accurate results with a Biosense breath meter.

[ Get $20.00 off a Biosense Breath Ketone Meter Here. ]

In the US, we generally measure blood sugar in mg/dL and blood ketones in mmol/L so we have to do a little conversion and divide the blood sugar reading by 18. In the UK or other countries where blood sugar is measured in mmol/L, it can be calculated as a simple ratio.

What do GKI results mean?

For therapeutic ketosis, the lower the GKI reading, the better. My goal is to stay under 1.0 for the majority of the day although some weeks I’ll relax and be satisfied with anything under 3.0. I’ve noticed that as I’ve become more fat adapted over time, it’s harder to stay in the sub-unity range without extended fasting. More on that below.

For folks pursuing nutritional ketosis for its general health benefits, anything under 9.0 is a good target.

What GKI numbers mean

9 or aboveNot in ketosis.N/A. Just living a normal life.
6-9Low level of ketosis.For weight loss or health maintenance.
3-6Moderate level of ketosis.For managing Type 2 diabetes and obesity, insulin resistance, metabolic or endocrine disorders.
1-3High therapeutic level of ketosis.For those using keto therapeutically for the treatment of diseases such as cancer, epilepsy, Alzheimer’s, Parkinson’s, traumatic brain injury, etc.
less than 1The highest therapeutic level of ketosisVery difficult to achieve without a doctor’s supervision.

Common factors that impact your GKI (Glucose-Ketone Index) include food and exercise.

Most common factors that impact your Glucose-Ketone Index (GKI)

Amount of carbohydrate consumed

This one is obvious. Carbohydrates will raise your blood sugar. The first law of keto club is to reduce carbs. Your body won’t start breaking down fat to create ketones unless it’s running low on glucose. The general rule of thumb is to eat fewer than 20g net carbohydrates in order to enter this ketone-generating state known as ketosis. Net carbs are total carbs minus fiber and sugar alcohols as, supposedly, the latter don’t impact blood glucose.

How much you actually have to reduce carbs before you start creating ketones, however, depends on your individual body and the type of carbs you’re consuming, as below. GKI tracking can help you understand the right amount for you.

Quality of carbohydrate consumed

Not all carbs are the same when it comes to the metabolic impact. For many people, including myself, processed and refined carbohydrates have a much larger impact on GKI than those from whole vegetable sources. For example, I was able to maintain a GKI below 1.0 for months eating 45 or more grams of net carbs per day in the form of broccoli, kale, cabbage, and other vegetables (but, let’s be honest, mostly broccoli, kale and cabbage). On the other hand, if I cheat and eat a Quest bar with 4g net carbs, or some other low carb processed food – even if it’s the only thing I eat – my GKI will immediately spike and it usually takes me a day or so of fasting to get it back down.

I’ve read dozens of accounts from people whose experience mirrors my own in this regard. It probably isn’t an issue for most people pursuing nutritional ketosis but, if you find you’re not able to enter ketosis even at 20g or fewer net carbs, try cutting out the processed food and artificial sweeteners. Tracking your GKI during days eating only whole foods versus processed foods can tell you how your body reacts.

Artificial sweeteners are incredibly processed. You’ve never marveled at the beautiful fields of maltitol or spent childhood summers hunting for allulose in the woods. While some are better than others and each may affect you differently, be aware that they will likely have an effect on your GKI no matter what the label says. Also, for many people including myself, artificial sweeteners will increase appetite leading to cravings and negating the big appetite-suppressing benefit of keto. Of course, the only way to know for sure how artificial sweeteners impact you is to test.

Protein consumed

In addition to dietary sugar sources, the liver creates glucose from protein using a process called gluconeogenesis. This could happen during periods of true starvation when the body runs out of fat to use as fuel and starts to break down muscle for energy. More often, this is the process the liver undertakes when more protein is consumed than is needed for immediate energy needs and the body plans to store it for later energy needs in the form of fat.

In other words, too much protein can increase blood sugar leading to an increase in GKI. Most casual keto-ers don’t need to worry too much about their protein intake and some are more sensitive to protein than others. Nevertheless, virtually everyone trying to get into deep therapeutic ketosis will need to monitor their protein intake as closely as their carbs. GKI tracking can be an enormous help in finding the right individual amount of protein to consume.

Intense Exercise

Those of us committed to fueling our body with fat via ketosis rather than glucose will notice that it works. That is, after intense exercise, blood ketones are usually lower because they’ve been used for energy. Lower blood ketones leads to a lower overall GKI but it will return to normal a couple of hours after exercise.

Even my mega chillax morning yoga practice usually lowers my blood ketones by .2-.5 mmol/L and increases my blood glucose by 5-10 ml/dL despite my not consuming any food.

Unexpected factors that can impact your GKI (Glucose-Ketone Index) include stress, sickness, sleep, gastropareis and adaptive glucose sparing.

Unexpected factors that can impact your Glucose-Ketone Index (GKI)

Stress and Conflict

This is surprisingly one of the biggest factors in my GKI monitoring now that I don’t have much variation in my diet. It’s part of the reason my mindfulness practices, yoga and mediation are so critical. A fight with Brad will send my blood glucose soaring 20 points or more and for weeks my morning GKI was elevated due to work stress.

It’s not just me. Diabetics have long been advised by medical authorities to monitor their stress levels to control blood sugar. Stress reducing activities like meditation, yoga, mindfulness practice, exercise, belly breathing and adequate sleep are all important parts of metabolic therapy and overall metabolic health.


Illness is a variation of stress, discussed above, and will affect your blood sugar in the same way. Additionally, if you’re dehydrated due to your illness the impact can be even more profound. Higher blood sugar means higher GKI reading even if you’re off food so drink plenty of water, rest, and don’t worry about your numbers until you feel well.


I forget to include this one all too often because sleep has never been a problem for this girl. However, those of you who aren’t prioritizing 8 hours a night, aren’t observing good sleep hygiene or otherwise struggle to be fully rested will see the consequences in your GKI.


Though not common, gastroparesis, or delayed gastric emptying, can cause mysteriously high blood glucose readings until you figure out the cause. Delayed gastric emptying means the food you eat just sits in your stomach long, long after it should have moved along the digestive tract. It can be caused by inflammation, diabetes, neuropathy, some cancers and cancer treatments or other conditions that might damage the vagus nerve like MS or Parkinson’s. It’s a total bummer.

There are lots of fancy ways to diagnose it with trackers you ingest, etc. Simpler, just find yourself vomiting up the green salad you ate 2 days ago, still green, or carrying a food baby completely to term. I honest to God no exaggeration went out and bought a pregnancy test.

What’s worse, the treatment protocol for delayed gastric emptying recommends a diet low in fat and fiber with more frequent meals, completely the opposite of my intermittent fasting, plant-based ketogenic way of eating. I mean, all I eat is fat and fiber. And I prefer to eat it all in one sitting a day. Ultimately, I’ve managed to find tentative balance with smaller meals eaten more frequently between longer periods of fasting.

Adaptive Glucose Sparing

Adaptive glucose sparing is as much of a bummer for those pursuing therapeutic ketosis as gastroparesis. However, if you’ve noticed an unexplained rise in fasting blood glucose it could be the culprit. I intend to do a full post on this. In the meantime, the quick summary is that if you’ve been in deep ketosis for a long time, your muscles may come to prefer fat as the primary fuel and refuse available glucose in the bloodstream. Even if you eat a 0 carb diet your liver will still be making glucose via gluconeogenesis and, with no organs willing to use it as fuel, it sticks around in the bloodstream leading to higher blood sugar readings and thus higher GKI. This isn’t really a problem for most but for those of us using keto for cancer I imagine all that unused glucose goes straight to fueling metastasis. Not that dire, I’m sure, but still.

When and how often should you test your GKI (Glucose-Ketone Index)

How often and when should you test your GKI?

This really depends on you and your goals. For me, there are three types of tracking that I find helpful: throughout select days, daily, and around individual events.

Track GKI Throughout the Day

I think that, for a few days at least, it’s helpful to test every 2-3 hours to develop an understanding of your natural patterns.

Graph of Maggie's Glucose-Ketone Index (GKI) througout the day

Each line in this graph represents my GKI over the course of a day from waking on the left to bedtime on the right. This type of tracking on “good” days and “bad,” fasting and feasting, helped me to identify normal and abnormal patterns over time. For example, the aqua line represents my second day water fasting when I’d expect to see something more like the orange or purple line. The fact that my GKI was so elevated despite not eating anything (and actually introducing metformin, a blood sugar reducing drug, to my regime) eventually led me to learn about gastroparesis and adaptive glucose sparing.

Daily GKI

Additionally, I like to keep a daily pulse on my GKI with a reading at the same time each day. For me, this is a few hours after waking when, as you can see from the chart above, my fasting GKI is at its highest. I also often include an evening measure prior to eating. Sometimes I’m strict about twice a day measures, especially when I travel and am introducing new foods. Other times I may go days without testing when I’m eating my usual diet and know the impacts.

Maggie's blood glucose ketone index over time.

You can see in this chart of my GKI over time how much simply knowing my numbers helped me to recalibrate and improve them when I first started tracking.

This type of testing is especially helpful in identifying and understanding the impact of disturbed sleep or regular exercise on your metabolic health. Exercise may temporarily spike your GKI in the short-term but lower it over time.

In the chart above you can see that daily tracking of my GKI (blue) revealed an issue much more clearly than just blood glucose alone (red). GKI clearly spiked and became elevated in August while the corresponding rise in blood glucose was less noticeable.

GKI Testing of Food and Activities

Finally, I find it enormously helpful to test the GKI impact of certain food and activities. To do this, I begin in a fasted state – at least 3 hours after eating although I usually start a minimum of 16 hours after eating.

Measure your GKI immediately before consuming the food or engaging in the activity in question, then at 30 minute intervals for the next 2-3 hours to see how your blood glucose and ketones respond. This is great for identifying sensitivities to certain foods like artificial sweeteners or processed treats. It can also uncover your body’s individual response to exercise, meditation, etc.

A definitive guide on how to leverage GKI tracking for therapeutic ketosis: why track, when to test, and factors that can impact GKI readings.

Final thoughts on GKI tracking for cancer and other therapeutic applications

Do it. If you don’t measure, you don’t know. I really don’t think its possible to effectively leverage ketosis for therapeutic purposes without tracking. There are so many personal factors that can affect your numbers its worth the small investment in your health to get a meter and test yourself regularly until you find your rhythm.

As the benefits of therapeutic ketosis are being scientifically demonstrated for more and more types of cancer, I don’t think it will be many more years before this is a standard part of cancer treatment testing.

You may also like


Colette Ellis December 7, 2019 - 6:28 am

This is really interesting, thank you. I have a glucose meter I will buy a ketogenic one. Do you know if metastatic brain tumours are the same as lung to brain mets or glioblastoma tumors? I gave MBC with mets to spine, liver, eye and brain. I don’t know if I should be doing ketogenic?

Maggie Jones December 7, 2019 - 3:10 pm

Hi Collette! Not every cancer benefits from keto but breast cancer is one that seems VERY responsive. A study published in just September 2019 (this is all so new) compared two groups of 30 women each with late stage breast cancer requiring surgery. One group was treated with chemo only before the surgery, the other with chemo + keto. 30 months later, the keto+chemo group had 100% survival compared to 60% of chemo only. This was the first larger scale human trial but reaffirms animal studies and n=1 anecdotal evidence.

seraphim August 6, 2020 - 10:21 pm

Dear Maggie,

Thank you for this wonderful site. I have type 2 diabetes and was diagnosed with a 2 cm tumor invasive ductal carcinoma. I chose to go alternative with cancer and am alive and well after they said I would be dead by now (a year later). I test glucose & ketones for GKI and exercise twice daily.

I have a question that I wish could please be answered: Is therapeutic level GKI (less than 3) better than moderate level GKI (3-6) for type 2 diabetes? It is very easy for me to be in a therapeutic level, but I don’t have cancer anymore and am addressing type 2 diabetes.

Thank you in advance for the help!


Maggie Jones August 7, 2020 - 8:23 pm

Seraphim, thank you so much for your kind words and for sharing your incredible story of healing. You are truly a miracle!

After crushing cancer, I usually advise a whole foods, plant-based diet incorporating more grains, legumes and fruits than are normally eaten on a therapeutic keto diet. It’s hard to advise on your particular case without knowing more details (let me know if you’d like to consult) but, if it’s easy to stay below 3.0 without incurring nutritional deficiencies, go for it!! You are an inspiration!

Laura September 1, 2020 - 4:31 am

Have you noticed your GKI not being as favorable just before or during your period? I’m wrapping up a 36 hour fast and my numbers are in low level ketosis range, which is not normally the case for me. I don’t typically do any fasts longer than 18 hours on my period (or in the week before it) so I’m wondering if this could be the difference for me.

Maggie Jones September 1, 2020 - 7:27 am

Holy smokes, Laura! Thank you so much! I cannot believe I didn’t address this here. Blood sugar and GKI are almost always higher immediately before and during your period, depending on the person, according to multiple diabetes research nonprofits. I happen to be expecting my period in the next day or two and am 36 hours fasted. I just checked my blood after reading this comment and indeed, my blood sugar is more than 10 points higher than usual at this point in the day and ketones are almost 1.0 mmol/L lower than I’d normally expect after a fast this long.

Honestly, I probably would have been pretty annoyed if I’d taken my GKI before getting your comment. Thank you so much for pointing this out! I’ll be updating the article to note how hormonal fluctuations can impact the GKI.

Congrats on your fast, too!

Bryce Marshall September 4, 2020 - 6:52 pm

This is super helpful! Thank you! In addition to GKI I’m also curious with around what level your glucose is when you’re aiming for killing cancer? Because my GKI is good because my ketones are super high(4-6mmol/L) but my glucose is barely lower than when I wasn’t on keto and if lack of glucose is what starves tumors, then it seems like that might be a problem. Thanks!

Maggie Jones September 5, 2020 - 2:34 pm

Hi Bryce! That’s a great question and, if your ketones are in the 4-6 range, you almost certainly have a GKI around 1. A good reason to focus on the ratio rather than your glucose number is that you have less control over glucose – even if you ingested zero glucose, your liver will continue to create it using gluconeogenesis. Whatever your glucose, the ratio will provide insight into your metabolic health. For reference, my blood glucose most days is 84mg/dL – as high as 90s or low 100s after conflict or when sick, as low as 50s after days of fasting (but it can also be 80s during fasting). So really, I stay focused on the ratio as the true indicator of metabolic health and use the glucose numbers to help me troubleshoot any changes there.

Hope that is helpful!


Leave a Comment

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More