#246

Or just do both and increase your lithium dose to compensate. (Even more concise) :wink:

1 Like
#247

This only works if it appears that lithium is really useful. What I pointed to in my original message ( Lithium Supplementation - #236 by adssx ) is that: evidence for the benefits of SGLT2i (hundreds of trials, longitudinal studies, animal models in different populations + ITP life extension) is incredibly more robust than for lithium. Despite SGLT2i lowering serum lithium levels. So the most likely outcome is that lithium is useless. It’s not what I believe, I think lithium supplementation is probably useful (but which lithium form? which dose?), but let’s be honest: the most likely probability is null effect (that’s the baseline for any intervention anyway).

3 Likes
#248

I can definitely tell you that Empagliflozin has an amazing effect. I have never drunk so much water in my life! It also seems to be a very good way to lose weight.

1 Like
#249

It has to be noted though, that all association studies on SGLT2i and dementia (and that’s all we have) are hopelessly confounded by one fact: long term data is in diabetic patients.

Diabetes raises the risk of dementia. Any treatment that improves the diabetic condition of the intervention group, also improves the hazard ration for all increased disease risks. So there really is not yet a strong case - unless you demonstrate, that diabetics taking SGLT2i have a lower dementia risk than demographically matched non-diabetics. Does this data exist?

Lithium does have this data. It’s a common drug for bipolar disorder. A condition that raises dementia risk as well. In recent studies, bipolar people using Lithium were associated with lower dementia risk than the general population.

However, none of this matters. The LATTICE trial is going to conclude by early 2025. It’s a randomized clinical trial of Lithium for prevention of dementia in people with early cognitive decline. And it’s powered sufficiently to give an answer - positive or negative.

3 Likes
#250

Since we are all subject to some degree of cognitive decline, the LATTICE trial results should be interesting. However, in the meantime, I see no reason why anyone should not take an SGLT2I and lithium. It sounds like they work through different pathways. If I’m wrong, you may be pissing away $.05 daily for your lithium supplements. Not a big loss.

2 Likes
#251

We also have two small trials + animal models. And we might have association studies in people without T2D but with HF or CKD but I cannot find any.

Trials:

Diabetes raises the risk of dementia but it’s protective in PD and yet for PD the association studies still show protection.

Diabetes raises the risk factor for depression and yet diabetics taking SGLT2i might have a lower depression risk than demographically matched non-diabetics: “In a post-hoc analysis using the non-diabetes population as reference, patients with diabetes using SGLT2 had even lower odds of depression compared to individuals without diabetes (ORSGLT2_non-diab 0.77 (0.61–0.98)).” (Diabetes, antidiabetic medications and risk of depression – A population-based cohort and nested case-control study 2022)

(I posted the sources for all the above at some point in the SGLT2 thread at some point)

4 Likes
#252

Last month, the Lancet commission on dementia published its 2024 report on dementia prevention, intervention, and care.

They did not include bipolar disorder in the list of modifiable risk factors: Dementia Prevention - 2024 report of the Lancet standing Commission

Still they noted:

Few studies have investigated dementia risk in people with bipolar disorder, but findings have been consistent regarding an increased risk.
There was heterogeneity in the extent to which studies adjusted for factors such as cardiovascular risk, comorbidities, and alcohol consumption.

Regarding lithium in BD, I found: Risk of dementia in bipolar disorder and the interplay of lithium: a systematic review and meta-analyses 2020

BD increases the risk of dementia (odds ratio (OR): 2.96 [95% CI: 2.09–4.18], P < 0.001), and treatment with lithium decreases the risk of dementia in BD (OR: 0.51 [95% CI: 0.36–0.72], P < 0.0001).

Bipolar people using lithium have a risk of dementia of 2.96 x 0.51 = 1.51 vs the general population? So not lower @Guest?

1 Like
#253

I assume you’re talking about this trial: Lithium As a Treatment to Prevent Impairment of Cognition in Elders (LATTICE)

Just completed last month! I emailed the PI to know when the results will be pubished.

6 Likes
#254

If SGLT2i lowers the risk of depression and Parkinson’s below the population average in diabetics… …but all the date we have so far show that it’s not the same for dementia… …this is not a strong case for the latter indication.

Your meta-analysis is problematic:
as you can see on page 7/516, the 3 most recent studies of the 6 studies used, DO SHOW that lithium is protective for dementia risk compared to the general population. Why is that relevant: the earlier 3 studies are tiny in comparison. It’s just 33, 114 and 220 controls and only a handful of lithium users. The more recent 3 studies have thousands to tens of thousands of participants. But the meta-analysis gives equal weight to all 6 studies !

For lithium I’m also referencing (using propensity scoring - a very close demographic matching algorithm to match statistical “twins”):

finding a hazard ration of 0,56 of lithium users vs. non-users in the general population.

Just to be clear: I’m not saying SGLT2i aren’t going to be protective against dementia in otherwise healthy people. I’m not saying everyone should take lithium to prevent dementia. But you seem to make the case, that we have sufficient data to make those recommendations. Which is simply untrue.

Again: in particular everything for SGLT2i is mixed up by the co-disease if diabetes. To make a point using association studies alone, you have to find studies that lower the risk below the population average.

As for lithium: let’s just wait the analysis of the patient data of the RCT !

5 Likes
#255

I don’t know if previous papers have looked at the general population level. This one looked at something close to that and found a good effect: Assessment of sodium-glucose cotransporter 2 inhibitors (SGLT2i) and other antidiabetic agents in Alzheimer’s disease: A population-based study 2024

By the way, regarding T2D and dementia, the Lancet commission 2024 report says: “New evidence suggests that age of onset makes a difference, with midlife, but not necessarily late-life, diabetes onset increasing the risk of dementia. […] Diabetes should be classified as a midlife risk for dementia. It is unclear whether diabetes is not a risk factor for dementia at older ages or whether the absence of evidence showing significant risk is because of short follow-ups and few studies. WHO concludes that diabetes in late life might have a detrimental effect on brain health and dementia risk.”

So, depending on the age of the analyzed T2D population, they might not even be at an increased risk of dementia. When people of various ages are grouped together, the HR of diabetes for dementia seems to be 1.82 (Associations between depression and cognition, mild cognitive impairment and dementia in persons with diabetes mellitus: A systematic review and meta-analysis 2022). Most association studies found HR of ~0.5 for SGLT2 use among people with T2D. So you could assume an overall HR vs the general population of 1.82 x 0.5 = 0.91. Just a rule of thumb, but it’s better than the similar rule of thumb for lithium I cited above (1.51).

Also, all SGLT2i are usually grouped together. There might be intra-class differences. For instance here empagliflozin and dapagliflozin were associated with a lower risk but not canagliflozin: Association of Sodium–Glucose Cotransporter 2 Inhibitors With Time to Dementia: A Population-Based Cohort Study 2022

It’s not “my” meta-analysis :wink: Please be hard on the science, not on the people :hugs:

Not exactly. As the authors note:

Caution should be exercised when drawing conclusions from the current study with regard to the general population. Our cohort differs from the general population, as our database was of patients treated for MH conditions, although the age structure of our database was similar to that of England. The frequency of dementia in our control cohort was higher than in the general population, as would be expected for a MH service.

I did not make this case; I don’t make any recommendations. I do hope that SLGT2i (and lithium and any other compound on earth!) will prove to be neuroprotective. I believe in lithium so much that I sponsored this experiment: Lithium Supplementation - #226 by adssx. I also donate to non-profits that run lithium RCTs in PD. But, unfortunately, we don’t know at this stage. That’s precisely why I started this discussion ( Lithium Supplementation - #236 by adssx ): we have two drugs that are touted as neuroprotective and yet “cancel” each other: that looked weird to me. (Although, as others said, there are many potential explanations.) Still, I think that the case for SGLT2i is stronger. Time will tell…

Yes, I’ll let you know if the PI answers my email! :pray:

4 Likes
#256

He answered literally one minute ago:

Yes – final study assessment was 8/1/24! We’re now undertaking data analysis. This is taking some time. We intend to submit our findings for publication by the end of the year. Thank you for your interest, Ari Gildengers

By the way (general question): if the data analysis is taking a lot of time: does it mean that the results are not obviously positive? I’m always surprised by how long it takes between the end of a trial and the publication of the results and I wonder if it’s because researchers are slow (they don’t have a massive incentive to publish a few months earlier) or because they’re looking at the data through every possible angle to find at least one positive result in one subgroup and avoid publishing a null result.

5 Likes
#257

This entirely depends on the field of science and common practices. For example in physics, results of major experiments (those requiring collaboration of hundreds of scientists) are announced in a timely fashion - long before publication. You might argue it’s unavoidable anyway, due to the number of potential leaks. In medicine it frequently takes months before results are announced, even for major RCTs with tens of thousands of participants and pressure by media and patient groups.

The hypothesis of the LATTICE trial is a large impact of lithium on dementia conversion for people with MCI. Once unblinding is done, the lead investigator privately know, if there is such a large effect or not. You don’t need to torture the data for that. But it’s a small trial with little media attention and it can be beneficial to keep a lit on the results to better “sell” the study to a journal for publication. It’s not an IP intervention (so difficult to acquire funding, as no industry money), it appears it’s mostly one investigator doing the heavy lifting, it took 7 years to plan and conduct the trial. So it’s all the more important for the author to place it in a well ranked journal to make the efforts worth it.

3 Likes
#258

We/I do a lot of things based on this premise :slight_smile:

4 Likes
#259

If telmisartan increases lithium levels, does it make sense to take telmisartan and lithium together? I take telmisartan in the evening and dapagliflozin in the morning so I chose to take my 1 mg lithium in the evening, but would it really make a difference?

4 Likes
#260

If telmisartan increases lithium levels, then the hypothesis that lithium might be neurotoxic (and so the SGLT2i is partially neuroprotective by the mechanism of getting rid of the toxic lithium), would mean that telmisartan elevates neurotoxicity by increasing neurotoxic lithium levels. Is there any evidence that telmisartan causes neurotoxicity? That could tell us something about lithium vs neural effects. Also, if telmisartan increases lithium levels, then those on high lithium doses should avoid telmisartan on account of high lithium levels being bad for the kidneys? Of course, telmisartan is kidney protective on account of lowering BP.

1 Like
#261

Many papers suggest that telmisartan might be neuroprotective. We don’t know how much telmisartan increases lithium concentration, but even if that’s 2x, it might just be too low for any effect to be seen based on the dietary lithium content only. In any case, I don’t think anyone has ever suggested that low-dose lithium might be neurotoxic. (My pure guess is that lithium is beneficial but at doses way above drinking water levels, somewhere between 1 mg and 50 mg?)

People on high lithium dose usually monitor their serum lithium levels and then adjust their lithium dose if they use telmisartan.

4 Likes
#262

Lithium for Parkinson’s: an Extension Trial

Because our earlier study showed maximum improvements in PD biomarkers in patients with serum lithium levels of 0.25-0.50mmol/L and there are large interpatient variations in serum lithium levels achieved from the same lithium dosage, this present study will adjust lithium dosing in each patient to achieve this target serum lithium level for an additional 24 weeks.

0.25-0.50mmol/L means the same in mEq/L. However, below 0.5 mEq/L is considered the subtherapeutic dose so I assume most labs can’t detect that.

2 Likes
#263

I tend to find that my lab can detect down to 0.05 mmol/L.

1 Like
#264

I’m glad to be wrong on this! Great news.

2 Likes
#265

I am sure labs will vary, however.

My problem is I would like to target around 0.025 mmol/L

1 Like