Biologic plausibility
Given the inherent inability of observational studies to determine causality, it becomes necessary to critically examine the observed associations between higher serum phosphorus and adverse outcomes. First and foremost, we need to determine if there are any plausible biologic mechanisms that could explain how higher serum phosphorus can cause vascular and/or renal damage.
Indeed, experimental models have shown that vascular or parenchymal calcification can be induced by an increase in ambient phosphorus levels (Circ Res. 2000;87:E10-E17). Other potential mechanisms whereby higher phosphorus could be damaging include a deleterious effect of secondary hyperparathyroidism and increased FGF-23 levels, hemodynamic alterations, and derangements in cellular energy metabolism (Am J Physiol 1993;264:F752-F759).
Experimentation
No matter how intellectually satisfying it is to understand how higher serum phosphorus may be detrimental, the clinical benefit of lowering phosphorus levels has to be proven in clinical trial settings. Unfortunately, as with many other interventions in CKD, there are very few clinical trials of phosphate-lowering therapy to support the benefit of such interventions.
The majority of the clinical trials in this field have been concerned with proving the superiority of one type of phosphorus-lowering medication over another. Such trials cannot offer insight into the benefit of phosphorus lowering, as they usually use agents with comparable hypophosphatemic effects.
There have been no placebo-controlled clinical trials of phosphorus lowering with mortality as an end point. Studies utilizing other clinically relevant end points are also few and small in size. Dietary restriction of phosphorus resulted in attenuated progression of CKD in two small clinical trials (Kidney Int Suppl. 1983;16:S278-S284 and Kidney Int. 1982;22:371-376).
However, the size of these trials, the inability to dissociate protein restriction from phosphorus restriction, and the lower BPs seen in patients whose phosphorus intake was restricted make it difficult to use these data in support of phosphorus-lowering therapies.
Better evidence was provided by a recent small randomized, controlled trial of two different phosphate binders vs. dietary phosphate restriction in patients with NDD-CKD that examined coronary calcification as an end point. In this study, the highest progression of coronary calcification was seen in the group treated with dietary restriction alone (Kidney Int. 2007;72:1255-1261).
Interestingly, serum phosphorus levels were similar in the different intervention arms of the study, which makes it somewhat difficult to associate the observed differences in coronary calcification with higher serum phosphorus levels. Urinary phosphorus, however, was significantly higher in the dietary restriction group, showing that the latter intervention was less successful in alleviating the burden of phosphorus retention in patients with residual kidney function (making this a quasi-placebo-controlled study) and allowing for speculation that the homeostatic mechanisms involved in maintaining normal serum phosphorus levels in NDD-CKD may also play a role in vascular calcification.
Medications approved for clinical use need to have a favorable risk-benefit profile. This is certainly the case with phosphate-lowering therapies, as long as we consider medication-induced short-term adverse events on the risk side and only the lowering of serum phosphorus on the benefit side.
The proposition becomes less clear if we consider benefits imparted by the lower serum phosphorus levels (or lack thereof, mainly because of the absence of clinical trials substantiating such benefits) and if we factor in on the risk side the longer-term health effects of some of the phosphate-lowering therapies (the discussion of which is beyond the scope of this article). Furthermore, financial “risk” is not negligible with some therapies (Nephrol Dial Transplant. 2007;22:2867-2878), making a risk-benefit assessment somewhat more complicated.
Is treatment worthwhile?
The lack of clinical trials proving the long-term clinical benefits of hypophosphatemic therapies remains a sore point and does not allow for an unequivocal affirmative answer. On the flip side, though, the absence of such trials also means that lack of benefit (or harm) from phosphorus-lowering therapies has also never been proven. This, combined with the strong observational data suggesting detrimental effects from higher serum phosphorus and the sound biologic plausibility of such observations, suggests that serum phosphorus lowering should continue to be implemented in clinical practice.
Hyperphosphatemia is one of the few metabolic abnormalities of CKD that have adequate corrective treatments. I believe implementing such treatments is desirable even if we lack the gold standard proof of clinical trials, provided that such treatments remain safe and do not pose an undue financial burden on the individual or the society.
Dr. Kovesdy is chief of nephrology at the Salem VA Medical Center, in Salem, Va.; assistant professor of clinical internal medicine at the University of Virginia, in Charlottesville; and a member of the Renal & Urology News Editorial Advisory Board.