Crucially-Important and Mind-Bending Interactions in Phosphate and Magnesium Homeostasis

The authors of these articles [Thumfart et al., 2008: (http://www.ncbi.nlm.nih.gov/pubmed/18701629); Wei et al., 2006: (http://www.pdiconnect.com/cgi/reprint/26/3/366.pdf)(http://www.ncbi.nlm.nih.gov/pubmed/16722031)] have discussed the evidence that serum magnesium (Mg) levels tend to be inversely correlated with serum parathyroid hormone (PTH) levels, and Wei et al. (2006) discussed all of the evidence that magnesium repletion can be protective against calcification and thrombosis and markers of cardiovascular disease. The inverse relationship between the serum Mg and PTH levels is not widely known in the literature, and I've only recently even seen research on it. It's really important, and Mg is just really important, in general, in my opinion. Most of the research and articles on Mg and PTH have focused on the hypocalcemic hypoparathyroidism that can occur in severe Mg deficiency, but repletion of Mg in severely-deficient animals or humans only increases serum calcium (Ca) back to normal levels, by increasing (restoring) the normal capacity of the parathyroid glands to release PTH in response to decreases in serum Ca. At serum Mg levels or dietary Mg supplies that are higher than those that are required for those most basic functions, Mg is thought to suppress PTH levels by acting as a "weak" activator, like a partial agonist, almost, of the calcium-sensing receptor(s) that mediate the suppression of PTH release in response to increases in serum Ca (Thumfart et al., 2008). Paradoxically, Mg can also increase urinary calcium excretion by reducing the reabsorption of calcium in the renal tubules (Thumfart et al., 2008). It's important to note that that increase in urinary Ca excretion would be likely to occur in conjunction with the decreases in the risk of nephrocalcinosis that researchers have generally found in response to increases in Mg intake. In the case of phosphate (Pi) repletion, the decreases in urinary Ca excretion are thought to result from the suppression of PTH-mediated bone resorption. Thus, Pi is thought to actually reduce the amount of Ca that is filtered in the glomeruli, and Mg may partially act by inhibiting Ca reabsorption in the renal tubules (proximal tubules and distal tubules). But the research I cited above suggests that Mg can increase the rate of urinary Ca excretion and even decrease the serum Ca levels and exert a concomitant, suppressive effect on PTH release. That's a really unusual set of effects. Pi can decrease serum Ca (an effect that is probably undesirable) and decrease urinary calcium excretion but can also elevate PTH levels, and that's an effect that could be attenuated, for better or worse, by an increase in Mg availability to the parathyroid glands or the Ca sensing proteins in the renal tubules, etc. Additionally, many of the bizarre derangements in the homeostatic regulation of Ca and Pi that have been found in response to long-term, excessive Pi supplementation could result, in some cases, from Mg depletion. I also get the general sense that a "high" Pi intake will tend to produce plasma volume expansion and lead to a reduction in urinary sodium excretion, and that could tend to oppose the natriuretic effect that high doses of Mg can sometimes produce. Another way of looking at it would be to say that a high or excessive Pi intake may produce plasma volume expansion by reducing Mg absorption or by increasing Mg turnover by other mechanisms, and Mg has sometimes produced low-level diuretic effects [Walker et al., 1998: (http://www.ncbi.nlm.nih.gov/pubmed/9861593?dopt=Abstract)], by mechanisms that aren't clear.

Incidentally, this is another article that includes a discussion of the antithrombotic effects that increases in Mg availability can produce [Maier et al., 2004: (http://www.ncbi.nlm.nih.gov/pubmed/15158909)], and I've been meaning to collect some of the articles that show the antithrombotic effects of Mg repletion or of elevations in the extracellular Mg levels (the steady-state, extracellular Mg levels are not necessarily or even usually going to be elevated much or at all, even in response to Mg supplementation that increases intracellular Mg levels).

What's really interesting is that low serum Mg and low serum Pi tend to go hand in hand and produce many of the same manifestations, including rhabdomyolysis and decreases in red blood cell (RBC) deformability and hemolytic anemia and decreases in RBC 2,3-DPG and ATP, etc. Oken et al. (1971) [Oken et al., 1971: (http://www.bloodjournal.org/cgi/reprint/38/4/468.pdf)(http://www.ncbi.nlm.nih.gov/pubmed/5571433)] found that Mg deficiency caused hemolytic anemia, reticulocytosis in combination with erythroid hyperplasia in the bone marrow (basically meaning that some erythroid colony-forming units in the bone marrow may be enlarged and hyperresponsive to erythropoietin and that the immature RBC's are more numerous and are also undergoing apoptosis at a high rate, because of Mg depletion), decreases in serum phosphorus (and, hence, serum Pi, also), and decreases in the RBC 2,3-DPG and ATP concentrations and in the overall glycolytic activity in RBC's. Piomelli et al. (1973) [Piomelli et al., 1973: (http://www.bloodjournal.org/cgi/reprint/41/3/451.pdf)(http://www.ncbi.nlm.nih.gov/pubmed/4690142)] also found hemolytic anemia in Mg depleted rats and cited research that had shown hypophosphatemia and hypomagnesemia to occur concomitantly in animals and humans.

It's also really important to note that a lot of articles have shown that Mg supplementation at dosages that would produce "desirable" effects, in my opinion, can decrease serum Pi or produce outright hypophosphatemia. And Pi supplementation can produce hypomagnesemia and intracellular Mg depletion. I think that the amounts of supplemental Mg that might be required to compensate for those effects of Pi repletion could be large and could be too high for many people to easily "accept." But, if the Mg is binding to Pi in the GI tract and precipitating, it's not going to be absorbed (there could be some solubilization in response to pH changes along the GI tract, but, for the most part, the precipitation is going to be permanent and is going to mean that the Mg and Pi are "lost"). One way of considering this would be to say that some percentage of a dose of Pi, from the diet or low-dose supplement, under a doctor's supervision, is going to be absorbed and some percentage is going to not be absorbed and probably bind some amounts of Mg and Ca. The intestinal absorption of Ca and the maintenance of serum Ca and renal Ca reabsorption are much more effectively maintained and regulated, in my opinion. Researchers have noted that the serum Ca is relatively stable, even in terms of circadian changes, than the serum Pi. The serum Pi fluctuates wildly throughout the day and in response to exercise, etc. I think the serum Mg levels are not as unstable as the serum Pi levels are, but the intracellular Mg concentrations are very easily depleted, such as in response to catecholaminergic stimulation, etc. Assuming one is using any supplemental Mg and Pi under a doctor's supervision, there's really a need to not be afraid to increase the supplemental intake of Mg, from Mg salts (such as magnesium hydroxide or magnesium oxide) slowly but relatively freely, to compensate for the reductions in absorption that are likely to result from the extra Pi. It wouldn't be a good idea to increase the Pi as freely as one might increase one's Mg intake, but part of the point of this is that an adequate degree of Mg availability is really obligatory for many of the effects of Pi repletion to be sustained in the longer term. One way to approach this type of problem would be to decide on some dosage of supplemental Mg that is tolerable and safe, under a doctor's supervision, and then to increase the ratio of Pi to Ca, assuming one would want to do this, in the first place. That would allow one to evaluate the effect of the Pi increase, from food or low-dose supplements, with the "knowledge" of the baseline effects that the initial Mg dosage produced. There's still a tendency for a lot of the research to focus on the most severe manifestations of the depletion of Mg or Pi or both (hypophosphatemia or hypomagnesemia), but intracellular Pi and Mg depletion tend to occur long before overt hypomagnesemia and hypophosphatemia occur. In any case, those articles I cited are just the tip of the iceberg. Resistance exercise that is done correctly, for example, can drastically deplete intracellular Mg and Pi concentrations, but the tendency has been to focus, in the case of Pi and RBC 2,3-DPG, on the short-term, post-exercise increases in RBC 2,3-DPG or serum Pi. But the more important issues have to do with the changes that occur in the days after the workout. It doesn't make sense to say that resistance exercise that correctly emphasizes the eccentric movement is going to increase RBC 2,3-DPG in the hours after exercise and then cause those levels and the intracellular Pi levels in skeletal or cardiac myocytes to also remain persistently elevated. Where would the Pi come from. It's likely that intense exercise can produce drastic depletions in intracellular Pi levels, but I haven't seen a lot of data on that. The increases in catecholaminergic transmission during resistance exercise would be expected to produce a significant depletion of intracellular Mg, and this has been shown to occur. And beta-adrenoreceptor agonists can produce hypophosphatemia and hypomagnesemia in the long term, etc. Even lowly L-methylfolate could reasonably be expected to increase Pi and Mg turnover, in my opinion, as a result of its apparent catecholaminergic effects.

Interactions of Phosphate and Calcium Homeostasis with the Coagulation Cascade: Potential Relevance to Depression and Other Psychiatric Symptoms

So the "bottom-line," "take-home" message of that last posting is that, in susceptible individuals or individuals in whom the coagulation cascade has been transiently or mildly activated by infectious mono or influenza, an increase in serum calcium within the normal range could produce depression or psychiatric symptoms by producing low-level thrombogenic effects (effects that essentially disturb mitochondrial functioning, as the feed-forward activation of the coagulation cascade essentially always does), and reducing serum calcium by reducing the dietary calcium or vitamin D intake could ameliorate those effects. Increasing the ratio of the phosphate to calcium intake could be a superior way of addressing those potentially-thrombogenic effects (and calcium influx promoting effects, in neurons) of increases in serum calcium. And idiosyncratic effects of glutamine supplementation might be addressed by decreasing the vitamin D or calcium intake or increasing the relative intake of phosphate, to some small extent, given the potential for slight "calcemic" and hypophosphatemic effects of glutamine. It's possible that an increase in serum phosphate would reduce calcium influx into platelets, given that increases in phosphate availability have reduced stimulus-induced intracellular calcium influx in beta-cells, for example, if memory serves (see past postings), and in other cell types. That's thought to be one mechanism underlying magnesium's antithrombotic effects (and relative absence of hemorrhagic effects).

I don't have time to go into the research, but, in my opinion, some of the research that would seem to rule out a role for the activation of the coagulation cascade in depression (http://scholar.google.com/scholar?q=coagulation+psychiatry&hl=en) does not rule it out, given that research in people with lupus and research on the coagulation cascade in general have shown that localized endothelial cell activation, such as in cerebral blood vessels, can occur and can cause localized microthrombi or low-level thromboses without producing measurable changes in the systemic coagulation parameters. Blood tests of coagulation parameters are notoriously insensitive and problematic, in my opinion. This is not a scientific statement, but, if it were possible to easily evaluate coagulation function, then monitoring people on warfarin wouldn't be so difficult and complex for both doctors and patients (the people taking warfarin, etc.), in my opinion. The coagulation cascade is extremely complex, and quantitative data on coagulation parameters are not going to tell one all that much about the individual and tissue-restricted effects of that state in any one person. The INR, for example, is very insensitive and displays a semi-logarithmic relationship with changes in the serum prothrombin levels, etc. I tend to think the ex vivo tests on platelet function are also not always going to have relevance to tissue-restricted (or endothelial-site-restricted) thrombogenic effects in the brain, for example. Benign intracranial hypertension/idiopathic intracranial hypertension commonly produces psychiatric symptoms, but that type of disease state may just be a slightly more extreme state along a spectrum of low-level thrombogenic changes that could potentially contribute to some forms of severe depression or chronic fatigue syndrome, etc. Those articles about visual dimming in depression could also indicate that low-level activation of the coagulation cascade is occurring, given the common occurrence of visual dimming in idiopathic intracranial hypertension (and the fact that idiopathic intracranial hypertension is thought to be not-infrequently caused, in part, by venous sinus thrombosis). Magnesium can also produce antithrombotic effects and may, in my opinion, be a lot less likely to cause bleeding, upon adjustment to a dose increase in 1-2 days, than many or most of the many other compounds that influence platelet activation and the coagulation cascade. In any case, it's worthwhile to remember that the use of Ginkgo biloba extracts has been associated with intracranial hemorrhages in many case reports (http://hardcorephysiologyfun.blogspot.com/2008/12/ginkgo-biloba-extracts-and-intracranial.html), and many compounds can reduce coagulation by mechanisms that could be very dangerous and unpredictable. So one would want to talk to one's doctor about this type of thing. Purines have produced antithrombotic effects in a lot of animal studies and appear to be a lot less likely to cause bleeding than most of these other physiological approaches, but that's just my opinion, based on my experiences during infectious mono, several years ago. These are all, obviously, just my opinions.

Is Every Side Effect Always Rare? What About Thrombocytopenia Induced by Some Forms of Heparin?

This is nothing that anyone doesn't already know, but every side effect is always described as being rare (http://scholar.google.com/scholar?q=%22is+a+rare%22+%22side+effect%22+OR+complication&hl=en&lr=). Is each of those 77,000+ side effects or complications always rare? This is just in relation to the last posting and the things I saw in it. I'm sorry to have to put this up on the blog, and it's deeply disturbing to me. But I'm going to put it up. This article is called "Finding Haystacks Full of Needles: From Opus to Osler" [Levine, 2005: (http://www.chestjournal.org/cgi/content/full/127/5/1488)(http://www.ncbi.nlm.nih.gov/pubmed/15888818)]. I'm not trying to make any kind of statement by posting this, and I know there's a daily need to use various forms of heparin. That's especially true in view of the absence of any substitute for it or source of more-reliably-nonimmunogenic heparan-based or heparin-like polymers. But the author discusses research estimating or finding evidence to indicate that there are 600,000 cases of thrombocytopenia associated with some forms of heparin (I'll call it "heparin-associated thrombocytopenia") each year. The condition is a form of autoimmunity and is driven by antibodies to different platelet proteins, etc. The research also estimates that 90,000 of those people die each year from it and that 300,000 of them experience thrombotic complications. The condition can go on for months and is very difficult to treat.

The author also notes that the condition is described as being "rare" in almost every instance in which it's discussed. It's fine to state that something is rare, if it is, in fact, rare or if one legitimately thinks it's rare (because of clinical experience, etc.). But one point of that article is that there's frequently a delay to the appearance of the devastating phase of the thrombogenicity, so that people may leave the hospital and may have been forced to discontinue the form of heparin (or may have not had any ill effects at all) and may then develop thromboses a month later. They may then go into the emergency room and be treated with heparin and deteriorate in a dramatic way, as one would expect. So some or many of the cases are probably never recognized. In any case, it's just one of those things. One could say that the researchers made inaccurate estimates of the deaths or complication rates, but I'll bet there's more research that's similar. I don't much feel like seeking it out, the way I wound up doing in the last posting.