Arterial calcification is a common problem in advanced kidney disease and contributes to the high prevalence of cardiovascular disease. There are two forms: neointimal calcification, associated with atherosclerosis, and medial calcification. The former is not exclusive to renal failure and occurs in anyone with atherosclerosis. It is unclear whether this has any clinical significance other than being a convenient marker of atherosclerosis. Medial calcification is independent of atherosclerosis and is strongly linked with chronic kidney disease (CKD). Recent data based on mammography show that there is a more than threefold risk of medial calcification in ESRD and that this risk may begin as early as stage 3 CKD.
Although disordered phosphate metabolism clearly plays a role in medial arterial calcification, it cannot by itself explain this problem, and strategies other than controlling hyperphosphatemia are needed. A large body of data implicates extracellular pyrophosphate (PPi), an endogenous inhibitor of hydroxyapatite formation, in arterial calcification. Humans lacking the ectoenzyme that produces PPi develop severe arterial calcification in childhood, and mice lacking the same enzyme also develop arterial calcification. Extracellular PPi may also be derived from intracellular PPi, and a mutation in the putative transporter (ANK) leads to ectopic calcification in mice, but primarily of joints rather than vessels.
Plasma levels of PPi are reduced in hemodialysis patients and correlate inversely with arterial calcification. This may be related to another key enzyme in extracellular PPi metabolism, tissue-nonspecific alkaline phosphatase (TNAP), which hydrolyzes PPi and induces arterial calcification when genetically overexpressed in vascular smooth muscle in vitro and in vivo. The activity of TNAP is increased in vessels from uremic rats, suggesting a pathologic role. Currently, little is known about the regulation of TNAP in vascular smooth muscle cells and why it is upregulated in renal failure.
Therapies based on pyrophosphate show promise as potential clinical tools. Both PPi and bisphosphonates (nonhydrolyzable analogs of PPi) inhibit arterial calcification in uremic rats, and recently developed small molecule inhibitors of TNAP can prevent arterial calcification in vitro. The doses of bisphosphonates required to inhibit vascular calcification in vitro are far greater than those used to inhibit bone resorption in humans. One potential drawback to this approach (and any potential therapy for ectopic calcification) is inhibition of bone mineralization, which requires a high local activity of TNAP to remove inhibitory PPi. Consequently, the nonhydrolyzable bisphosphonates, but not PPi, inhibit bone formation at doses required to prevent arterial calcification in rats.
Two other endogenous inhibitors of arterial calcification, matrix gla protein and osteopontin, also appear to act through direct inhibition of hydroxyapatite formation but probably do not play a primary pathogenic role in the vascular calcification of CKD, inasmuch as both are upregulated in vessels from uremic rats. Matrix gla protein requires vitamin K–dependent γ-carboxylation. Deficiency—either genetic or related to warfarin use—leads to vascular calcification in animals and humans. Osteopontin is, molecule for molecule, the most potent known inhibitor of hydroxyapatite formation, but deficiency does not lead to vascular calcification unless coupled with deficiency of another inhibitor. These proteins have limited therapeutic potential because matrix gla protein is extremely insoluble, and osteopontin has other inflammatory actions. However, vitamin K could be of benefit because patients with advanced renal failure may have vitamin K deficiency. Magnesium also inhibits hydroxyapatite formation and accounts for most of the inhibitory activity in plasma, but its therapeutic potential has not been explored.
Thiosulfate is another compound that can inhibit vascular calcification in vivo and in vitro and is often used to treat calciphylaxis. Although it is present endogenously, the levels are far below those required to inhibit calcification. It is widely assumed that thiosulfate acts by chelating calcium, but recent data indicate that its interaction with calcium ions is extremely weak and that there is no effect on hydroxyapatite formation or dissolution. Thus, its mechanism of action remains to be determined.
It is clear that the arterial wall has a propensity to calcify, even in the absence of altered mineral metabolism, and that endogenous inhibitors, particularly pyrophosphate, are required to prevent this. Thus, arterial calcification must be seen as a failure of these endogenous mechanisms. Although these inhibitors can be the basis for future preventive and therapeutic strategies, their unwanted effects on skeletal mineralization must also be considered.