Higher serum phosphate concentration and higher dietary phosphate consumption are associated with adverse health outcomes, including all-cause mortality, in CKD1,2 and the general population.3 In this issue of Kidney360, Williams et al. used a mouse model of CKD (X-linked Alport syndrome) and wild-type mice to compare effects of a high phosphate, Western-type diet with a low phosphate, vegetable protein diet on the heart and mineral metabolism markers.4 Interestingly, mice with X-linked Alport syndrome are resistant to vascular calcification, which permits evaluation of the two diets on the heart independent of vascular calcification. The investigators had previously shown that CKD in this model decreased cardiac mitochondrial respiration before the onset of left ventricular hypertrophy.5 In the present study, mice received either a Western-type casein-based, 1.2% phosphate with 0.6% calcium diet, herein referred to as the high phosphate diet, or a vegetable protein-based diet containing 0.61% total phosphate (0.33% of that being non–phytate-bound) with 0.8% calcium, herein referred to as the low phosphate diet.4 Protein content was the same between the two diets, and mice started the assigned diet at age 75 days. Mice fed the high phosphate diet were sacrificed at 190 days versus 225 days for mice fed the low phosphate diet because the investigators were concerned that most of the CKD mice fed high phosphate would die before 225 days. Because CKD severity in the X-linked Alport syndrome mice was heterogeneous, mice with CKD were selected for further study if they had a BUN > 43 mg/dl. In addition, because some wild-type mice had CKD reflected by a high BUN, wild-type mice were selected for further study if BUN was <35 mg/dl. This meant that mice that were subsequently studied had severe CKD in the CKD group or no CKD in the wild-type group. With these factors in mind, mice with CKD had predictable changes in mineral markers with the high phosphate diet. Specifically, they had high serum phosphate, intact parathyroid hormone (iPTH), and intact fibroblast growth factor-23 (iFGF-23) and low klotho protein levels in the kidney. CKD mice fed the low phosphate diet had lower phosphate and higher serum calcium, likely because this diet contained more calcium. Serum iPTH and iFGF-23 were somewhat lower on this diet, but not statistically significantly different from corresponding parameters observed in mice fed the high phosphate diet. Kidney klotho protein levels were also not meaningfully different in CKD mice on the different diets. In contrast to CKD mice, the two diets did not induce differences in serum phosphate or calcium concentrations in wild-type mice without CKD. However, serum iPTH and iFGF-23 were higher and kidney klotho protein levels were lower in wild-type mice fed the high phosphate diet. Perhaps the most striking finding of all was the two-fold higher kidney klotho protein levels in wild-type mice fed the low phosphate diet. The investigators also studied effects of these two diets on cardiac mitochondrial function in these mice. Wild-type mice fed the low phosphate diet had higher levels of oxidative phosphorylation, whereas CKD mice on the high phosphate diet had lower levels. Curiously, oxidative phosphorylation was similar between CKD mice on the low phosphate diet and wild-type mice on the high phosphate diet, suggesting that the high phosphate diet is sufficient to reduce cardiac mitochondrial function even in the absence of CKD. Some limitations of the study include differing exposure times for the two diet interventions and differing composition of protein and calcium content in the two diets, making it difficult to conclude with certainty that dietary phosphorus content is the key exposure accounting for these findings. Furthermore, serum vitamin D levels and soluble klotho were not quantified, and so, the pathophysiological mechanisms linking dietary phosphate to CKD-mineral and bone disorder must be inferred. It is likely that the high phosphate diet decreased klotho in the kidney through fibroblast growth factor-23 inhibition of 1,25-dihydroxyvitamin D production.6 These pieces of information would have filled in some gaps. An interesting finding was that the CKD mice on the low phosphate diet had higher BUN levels and kidney fibrosis scores than CKD mice fed the high phosphate diet. Because mice eating the low phosphate diet were studied at an older age (35 days older than the high phosphate diet), the investigators posited that the higher BUN and fibrosis scores could be explained by the older age at which mice fed the low phosphate diet were studied. While this might be the case, it is also revealing that CKD mice fed the low phosphate, vegetable protein diet had more kidney disease and fibrosis compared with CKD mice fed the high phosphate, casein protein diet. This finding goes against the idea that a vegetarian diet might limit CKD progression. Analyzing kidney histology and function at similar ages could have helped resolve this question. What are the clinical implications of these findings? For CKD, the novel finding is that a high phosphate diet impaired cardiac mitochondrial activity, even in the absence of established vascular disease, potentially establishing the preconditions that are necessary for development of left ventricular hypertrophy. Perhaps more striking are the observations made in mice without CKD, specifically, that higher consumption of phosphate may impair cardiac mitochondrial function. What remains to be determined, however, is whether reductions in mitochondrial respiration in isolated myocardial fibers from these mice translate into sustained left ventricular dysfunction affecting human health. Food sources of phosphate are well known to the nephrology community. What is concerning is the potential for higher consumption of phosphate from food additives.7 Over 240 phosphorus containing food additives are available in the United States, and more than half of products sold by the top 25 food and beverage manufacturers in the United States contained at least one phosphate additive.8 These phosphate additives provide no nutritional value. In contrast to phosphates contained in phytate, which have low bioavailability, phosphates in food additives are inorganic and highly bioavailable. Educating patients to avoid phosphate additives can mitigate hyperphosphatemia in patients with ESKD.9 While food manufacturers are required to list phosphate-containing additives on the ingredient label, they are not required to report the amount of added phosphate or total phosphate content of their products. Furthermore, not all additives contain phos in their name, which can make it challenging for patients to identify high phosphorus foods. Examples include lecithin or modified corn starch. The findings from Williams et al.4 add to the growing body of literature supporting plant-based diets, which contain less bioavailable phosphorus, in CKD.10 Perhaps more importantly, they raise concern about the potential harms of excess phosphate consumption in the general population. Given the abundance and high bioavailability of phosphate-containing food additives, more research is needed to determine whether consuming too much phosphate is bad for health in the general population.
Pao et al. (Fri,) studied this question.