Those diseases may well be linked, but in reverse order. Vitamin D deficiency is caused by:
1. Reduced gut uptake of (fat-soluble) Vitamin D due to malabsorption.
2. Reduced renal hydroxylation of 25-hydroxyvitamin D3 into 1,25-dihydroxyvitamin D3.
I can't speak to your daughter's specific problems, but since she has gut and, possibly, renal disease (given her diabetes), Vitamin D deficiency is quite possibly a consequence of those problems, and almost certainly not cause.
I can't speak to your daughter's specific problems, but since she has gut and, possibly, renal disease (given her diabetes)...
The complications of chronic hyperglycaemia take years to develop, and T1D develops fast. There's no way a newly diagnosed T1D would have renal complications at the time of diagnosis, never mind a year earlier.
I haven't seen any human data on this, so please share if you have, but in animal models of T1D, impaired calcitriol conversion occurs within days of the onset of diabetes. If they had to develop renal failure in order to have impaired Vitamin D metabolism, then I would agree that the timecourse makes no sense.
The rising incidence of Type 1 diabetes may be due, in part, to the current practice of protecting the young from sun exposure. When newborn infants in Finland were given 2,000 international units a day, Type 1 diabetes fell by 88 percent, Dr. Holick said.
So I suspect that your statement, "vitamin D deficiency is ... of those problems [including type-1 diabetes mellitus] ... almost certainly not cause," is undermined by the evidence. Do you know of some problem with the study cited in the article?
Thank you for disagreeing. The article was actually so sparse with the information that I misunderstood it for an epidemiological association. I went and found the study in pubmed, and I see that it's actually stronger than that - it's a cohort study with an intervention. It's not gold-standard, randomized-trial-level evidence, but it's still important.
I'm still a bit skeptical, because the human genetic data, so far, doesn't back up the association between Vit D deficiency genes and T1D. Nevertheless, I'll revise my statement to agree that both (a) diabetes can cause a decrease in the hormonally active form of vitamin D, and (b) high-quality Grade B evidence suggests that the converse is also true.
Nevertheless, I don't see Vitamin D deficiency as the unifying diagnosis here. One could even speculate that the Vit D deficiency was secondary, and the T1D was tertiary, but we really don't have enough information to speculate here and I'll stop now. I probably shouldn't have in the first place.
Your point is well-taken, although it's one step removed from the point I'm getting at. My point, it's true, assumes that more than 0% of vitamin D deficiency is genetic. In other words, my point only stands if the heritability of Vit D levels is greater than 0. From that assumption, I discuss the association between Vit D genetics and T1D. (The heritability of Vit D levels has actually been measured, and has been found to be very high - greater than 60% in some populations - so genetic factors are driving a huge amount of the variation that we see.)
If there were an association between T1D and Vit D, you would expect an excess of Vit-D-associated variants among those with T1D when compared to matched individuals. It is true that if 75% of the population is Vit D deficient and most of this is due to environmental effects, this would would hurt your power, requiring a larger study before convincing yourself that the null hypothesis really should not be rejected. Hence the couching of my language with "so far," since I know that associations can definitely be uncovered as sample size increases. But given the high heritability, at least in some populations, we shouldn't need a stratospheric sample size to detect association.
How much of the observed vitamin-D deficiency is caused by known Vit-D-associated genetic variants? You can measure heritability without knowing what alleles in particular code a trait, right?
Correct, you can measure heritability without any genetic information whatsoever. Because of this we can know that, say, 60% of Vit D variability is heritable.
In short, 4 base pairs (out of 3 billion) account for 2.8% of the population variance of circulating Vitamin D. With such a small # of people, it's not surprising that few independent SNPs hit the genome-wide significance level, but it's still pretty impressive that the 4 SNPs that they found can account for so much of the variance. (In other words, it looks like they identified the source of ~5% of the genetic variance, or 2.8% of the total variance.)
This looks much like the lipids story: early GWASs give you a handful of loci with influence, and later GWASs with may more participants will give you a much richer set of loci that collectively explain 20-30% of the variance. And keep in mind that these are just the common SNPs.
1. Reduced gut uptake of (fat-soluble) Vitamin D due to malabsorption.
2. Reduced renal hydroxylation of 25-hydroxyvitamin D3 into 1,25-dihydroxyvitamin D3.
I can't speak to your daughter's specific problems, but since she has gut and, possibly, renal disease (given her diabetes), Vitamin D deficiency is quite possibly a consequence of those problems, and almost certainly not cause.