3 ms·

This is really cool and I think you're doing more due diligence than any HN user I've ever seen. The confounding variable is that you lost a lot of weight, of

This is really cool and I think you're doing more due diligence than any HN user I've ever seen. The confounding variable is that you lost a lot of weight, of course, and I think many people who lose weight will be healthier at least in the short term. A question I've had in my head for a while is whether someone who has to stay on a ketogenic diet to be a healthy weight has "healed" or has just masked the underlying issue. Would be interested to see your thyroid numbers as you continue on your ketogenic/fasting diet. "In range" does not necessarily mean optimal, and in general I'd be worried about T3 uptake going from 33 to 23 even though both are in range. The liver converts T4 to T3 in response to glucose availability -- this happens in many tissues, but they all depend on a liver enzyme, and most of the conversion occurs in the liver. Low CO2 is a sign of kidney stress and acidosis. I also believe it's a marker of a low metabolic rate, much like thyroid. Prolactin is a longer discussion I don't want to get into here (much like you on your LDL-C :)) but I'll definitely be following your posts as I think you're doing your part and have done more research than I have so maybe I'll comment on your blog sometime. Put very briefly I believe high prolactin in men can be a sign of chronic stress and calcium deficiency. It's been linked to gynecomastia, for example, which makes a lot of sense. In general I like to look for tests that show whether something's wrong, rather than what's specifically wrong. Someone might have alright D but still have issues properly metabolizing calcium.

lhl 7y ago
I'd argue that weight loss isn't a confounding variable, as resolving dyslipidemia and MetS is actually one of the main points of the intervention. As for "masking the underlying issue" - let's face it, if you go back to regularly eating a terrible diet, yes, you'll become unhealthy again, but the same would go be true for any chronic insult. Personally, I'm a believer in metabolic flexibility and I do take time off (vacations, parties and special events, or just when I feel like it) to indulge. The difference now is that I am more acutely paying attention to how the food is making me feel (turns out for example when I have excess bread/sugar, I end up getting a pimple or two that I had originally never connected to being driven by dietary excesses), having a framework to think about how the food will affect my health (what do I think my liver glycogen levels are at?), and much pickier in general about how much I'll eat (my "worth it" taste bar is much higher - if it doesn't taste good I can now just stop). I also feel like I now have effective tools to pull various metabolic levers, whereas before I didn't. Regarding thyroid numbers, IMO taking about what is "optimal" is actually much trickier than "in range" (which are just standard deviations) because they need to be taken in the context of how you actually feel and if you present any symptoms (for hypothyroidism that would be fatigue, increased sensitivity to cold, weight gain, muscle weakness, etc) - which may be another reason people don't post numbers. Surrogate markers are most useful 1) sequentially per individual and 2) need to be considered with whether those numbers actually correlate with symptoms/reflect pathophysiology. Some are pretty obvious, but thyroid numbers seem to be one of those that requires a lot of careful nuance, but seems not worth the trouble (tracking TSH, T3, T4, rT3, rT4, TPO, Tg) unless you're presenting. For kidney issues, eGFR is something you'd look at, you'd want to take a look if any of your CMP numbers are out of wack, but I personally wouldn't be worried about acidosis - it's almost never a problem for non T1Ds (would be a sign of a major insulin:glucagon issue at least) but you'd definitely know (and need to know) if it were an issue well outside of regular bloodwork. If metabolic rate is a concern, I think RMR and DXA are a much better combo for tracking, btw. As for prolactin, if that's something you want to track and can use as a long term marker that is actionable for you, great, more power to you. It looks like it's only $60 out of pocket, so not terrible, but just recognize that if you throw it out there as something that a number that people should know or care about, or that you want to compare with, it simply won't be. For chronic stress, I'd suggest 24h cortisol, and maybe some inflammatory markers (I use hsCRP as my one catch all canary, but other common ones include ESR, IL6, TNFa), but for most people, you'd probably get more value day to day from mood and sleep tracking (especially ones that can track HRV or CNS values like the EMFIT). Calcium is measured via the CMP and deficiencies would I suppose show up in a DXA as well (eg if your bone density Z-score was quite low), but again, I think that's a thing where you'd only be worried about it if you were presenting symptoms (extreme fatigue, hair and nail issues, sleep issues, etc). I do think that if you are trying to track down nonspecific issues (eg "fatigue") you're going to have to do a bunch of detective work, which may involve more and more esoteric labs, but I'd also posit that most of those could probably also be addressed by relatively simple lifestyle interventions, starting with an appropriate/nutrient-dense diet and the most basic mood/energy tracking (eg:1-5 phone app like MySymptoms or Daylio correlated with interventions), and an engineering mindset.
- hoka-one-one 7y ago
  I think a lot of people are aware of prolactin because a lot of overweight men have manboobs and don't want them. But up to you, at least you're testing liver and cholesterol -- not everyone can say that. I listed it first because if someone would only agree to test one thing and nothing else, I think it should be prolactin. Well the suggestion wasn't for me personally, it was for people doing keto. My contention is that if any of those numbers are off, they're probably wrong or missing something about keto. Interesting pts on the thyroid stuff, I'll have to think about it
  - lhl 7y ago
    OK, so I haven't done much research on this particular issue, but lets say you have an issue with gynecomastia - this presents itself as a fairly obvious physical symptom. Does prolactin measurement change either your diagnosis or the intervention - eg, does it tell you anything you don't know beforehand or change how you would correct the issue? Personally, if we're talking about single markers, I think fasting insulin is the one I would pick because it does fulfill those criteria - it gives you a 10y+ lens before other markers do on diabetic pathophysiology (which now affects 80%+ of American adults) and also give you guidance on how to treat/how effective your interventions are. The negative is that there is no home test (yet, this is being worked on by some people), so it's not as good as something that can be measured more often/continously. Everyone who gets an annual physical/bloodwork should be getting their liver numbers (at least AST/ALT, although in the US GGT is not standard) with their CMP, although most people (clinicians too) don't really understand what they mean. I like this presentation on liver markers btw: A/Prof. Ken Sikaris - "Fatty Liver & Chemical Pathology." https://www.youtube.com/watch?v=2c_QYbCxC6g https://www.youtube.com/watch?v=2c_QYbCxC6g. It's also important to get these sequentially to see what the actually progression is (the time frames are usually not captured since people tend to get these tests once a year at most). A basic lipid panel is cheap and is somewhat useful for diagnosing gross dyslipidemia but it's worth noting that it's also much more mobile than most clinicians believe: Grant, Azure D., and Gary I. Wolf. “Free-Living Humans Cross Cardiovascular Disease Risk Categories Due to Daily Rhythms in Cholesterol and Triglycerides.” Journal of Circadian Rhythms 17, no. 1 (April 24, 2019): 3. https://doi.org/10.5334/jcr.178 https://doi.org/10.5334/jcr.178.

lhl 7y ago

I'd argue that weight loss isn't a confounding variable, as resolving dyslipidemia and MetS is actually one of the main points of the intervention. As for "masking the underlying issue" - let's face it, if you go back to regularly eating a terrible diet, yes, you'll become unhealthy again, but the same would go be true for any chronic insult. Personally, I'm a believer in metabolic flexibility and I do take time off (vacations, parties and special events, or just when I feel like it) to indulge. The difference now is that I am more acutely paying attention to how the food is making me feel (turns out for example when I have excess bread/sugar, I end up getting a pimple or two that I had originally never connected to being driven by dietary excesses), having a framework to think about how the food will affect my health (what do I think my liver glycogen levels are at?), and much pickier in general about how much I'll eat (my "worth it" taste bar is much higher - if it doesn't taste good I can now just stop). I also feel like I now have effective tools to pull various metabolic levers, whereas before I didn't. Regarding thyroid numbers, IMO taking about what is "optimal" is actually much trickier than "in range" (which are just standard deviations) because they need to be taken in the context of how you actually feel and if you present any symptoms (for hypothyroidism that would be fatigue, increased sensitivity to cold, weight gain, muscle weakness, etc) - which may be another reason people don't post numbers. Surrogate markers are most useful 1) sequentially per individual and 2) need to be considered with whether those numbers actually correlate with symptoms/reflect pathophysiology. Some are pretty obvious, but thyroid numbers seem to be one of those that requires a lot of careful nuance, but seems not worth the trouble (tracking TSH, T3, T4, rT3, rT4, TPO, Tg) unless you're presenting. For kidney issues, eGFR is something you'd look at, you'd want to take a look if any of your CMP numbers are out of wack, but I personally wouldn't be worried about acidosis - it's almost never a problem for non T1Ds (would be a sign of a major insulin:glucagon issue at least) but you'd definitely know (and need to know) if it were an issue well outside of regular bloodwork. If metabolic rate is a concern, I think RMR and DXA are a much better combo for tracking, btw. As for prolactin, if that's something you want to track and can use as a long term marker that is actionable for you, great, more power to you. It looks like it's only $60 out of pocket, so not terrible, but just recognize that if you throw it out there as something that a number that people should know or care about, or that you want to compare with, it simply won't be. For chronic stress, I'd suggest 24h cortisol, and maybe some inflammatory markers (I use hsCRP as my one catch all canary, but other common ones include ESR, IL6, TNFa), but for most people, you'd probably get more value day to day from mood and sleep tracking (especially ones that can track HRV or CNS values like the EMFIT). Calcium is measured via the CMP and deficiencies would I suppose show up in a DXA as well (eg if your bone density Z-score was quite low), but again, I think that's a thing where you'd only be worried about it if you were presenting symptoms (extreme fatigue, hair and nail issues, sleep issues, etc). I do think that if you are trying to track down nonspecific issues (eg "fatigue") you're going to have to do a bunch of detective work, which may involve more and more esoteric labs, but I'd also posit that most of those could probably also be addressed by relatively simple lifestyle interventions, starting with an appropriate/nutrient-dense diet and the most basic mood/energy tracking (eg:1-5 phone app like MySymptoms or Daylio correlated with interventions), and an engineering mindset.

hoka-one-one 7y ago
I think a lot of people are aware of prolactin because a lot of overweight men have manboobs and don't want them. But up to you, at least you're testing liver and cholesterol -- not everyone can say that. I listed it first because if someone would only agree to test one thing and nothing else, I think it should be prolactin. Well the suggestion wasn't for me personally, it was for people doing keto. My contention is that if any of those numbers are off, they're probably wrong or missing something about keto. Interesting pts on the thyroid stuff, I'll have to think about it
- lhl 7y ago
  OK, so I haven't done much research on this particular issue, but lets say you have an issue with gynecomastia - this presents itself as a fairly obvious physical symptom. Does prolactin measurement change either your diagnosis or the intervention - eg, does it tell you anything you don't know beforehand or change how you would correct the issue? Personally, if we're talking about single markers, I think fasting insulin is the one I would pick because it does fulfill those criteria - it gives you a 10y+ lens before other markers do on diabetic pathophysiology (which now affects 80%+ of American adults) and also give you guidance on how to treat/how effective your interventions are. The negative is that there is no home test (yet, this is being worked on by some people), so it's not as good as something that can be measured more often/continously. Everyone who gets an annual physical/bloodwork should be getting their liver numbers (at least AST/ALT, although in the US GGT is not standard) with their CMP, although most people (clinicians too) don't really understand what they mean. I like this presentation on liver markers btw: A/Prof. Ken Sikaris - "Fatty Liver & Chemical Pathology." https://www.youtube.com/watch?v=2c_QYbCxC6g https://www.youtube.com/watch?v=2c_QYbCxC6g. It's also important to get these sequentially to see what the actually progression is (the time frames are usually not captured since people tend to get these tests once a year at most). A basic lipid panel is cheap and is somewhat useful for diagnosing gross dyslipidemia but it's worth noting that it's also much more mobile than most clinicians believe: Grant, Azure D., and Gary I. Wolf. “Free-Living Humans Cross Cardiovascular Disease Risk Categories Due to Daily Rhythms in Cholesterol and Triglycerides.” Journal of Circadian Rhythms 17, no. 1 (April 24, 2019): 3. https://doi.org/10.5334/jcr.178 https://doi.org/10.5334/jcr.178.