Hacker News new | ask | show | jobs
by entitydc 2194 days ago
You’d be off base in questioning any of that, at least as the central premise that “X person didn’t have Y effects because they ate too much Z macro.”

Keto, when strictly followed, provides a great reduction to bolus needs (in personal experience, 80-90% reduction, depending on protein intake), and a lesser, but still significant, reduction to basal needs (30% in my case); however, both types of insulin are absolutely still needed to prevent high blood sugars in an environment with no insulin, which is a significant risk of DKA (and death). Protein metabolizes into excess sugar in the blood stream as well, so unless you’re eating a pure fat diet, you will still need insulin. In addition to that, there are hormones (cortisol for sure), and other non-macro nutrients and alkaloids (a common report is caffeine, but this very much depends on individual biochemistry) that can cause rises in blood sugar, which require insulin to compensate for.

A type 1 diabetic without honeymoon insulin production will still die after a period of even the strictest of ketogenic diets. It was the treatment prior to the introduction of insulin, and was effective at extending the lives of diabetics; they would typically be able to live up to 3 or 4 years post-diagnosis before eventually dying.

For the record, my body enters keto at or around 45g/day, as measured by the presence and amounts of ketones; anyone who says “you need exactly N grams per day or it isn’t keto” is extrapolating from generalized experience at best.

The problem with keto isn’t that it isn’t effective for some things, it’s that people go a long way to oversimplify both the science behind it and the impacts it can have on other conditions by over applying that lens of simplification. It’s a good diet for some, but it isn’t a free lunch (pun not intended) - and it’s far from something that can eliminate the needs of insulin in type 1 populations, as this article alludes to.