NKF: High-Normal Phosphate Levels Linked to Mortality Risk

By Charles Bankhead
DALLAS, April 7 -- Serum phosphate levels at the upper limit of normal are associated with a significant increase in mortality risk in patients with normal renal function, investigators said here.In an retrospective cohort study of more than 13,000 largely inner-city patients, the mortality risk rose by 24% when a serum phosphate level was higher than 3.6 mg/dL compared with a serum phosphate level lower than 3.0 mg/dL, said Matthew Abramowitz, M.D., of Albert Einstein College of Medicine in New York.Every 1.0 mg/dL increase in phosphate levels increased the mortality hazard ratio by 0.12, he reported at the National Kidney Foundation's clinical spring meetings.
"In this cohort, 99% of patients had serum phosphorus values of 4.5 or less, which is within the normal range," said Dr. Abramowitz. "Six patients had phosphorus levels greater than 5."
Elevated phosphate levels have been linked to an increased mortality risk in patients with end-stage renal disease, early-stage chronic kidney disease, and in the mostly white patients of the Framingham Heart Study. Whether the same is true of a mostly minority, inner-city population of patients with normal kidney function had not been studied, said Dr. Abramowitz.
Investigators evaluated serum phosphate levels measured in 13,092 patients seen at Montefiore Medical Center clinics from January 2000 through December 2002. All patients had normal renal function as reflected by an estimated glomerular filtration rate of greater than 60 mL/min/1.732.
In addition to serum phosphate values and eGFR, baseline data included serum creatinine, demographics, comorbidities, and insurance status. Outcomes were determined by linkage to Social Security records. Patients were followed for as long as five years.
After stratification by baseline serum phosphate levels, patients in the top quartile of phosphate values (>3.6 mg/dL) had a mortality hazard ratio of 1.24 compared with patients in the bottom quartile (<3.0 mg/dL). Analysis for trend across quartiles resulted in a statistically significant difference (95% CI 1.04 to 1.47, P=0.03).
The 12% increase in the mortality hazard ratio associated with each 1.0 mg/dL increase in serum phosphate also was statistically significant (95% CI 1.01 to 1.24).
Patients in the highest quartile of serum phosphate were younger, more likely to be female, non-Hispanic blacks, and to have diabetes. They had higher eGFR values and serum calcium levels and lower bicarbonate and hemoglobin levels.
Dr. Abramowitz emphasized that the study showed an association, not a causal link, between serum phosphorus levels and an increased mortality risk.
"Phosphate might simply be a marker of something else, such as dietary factors that we did not have enough data in our cohort to look at -- protein intake, soda intake," he said. "It's difficult to recommend any kind of treatment based on these findings.

Primary source: National Kidney FoundationSource reference:Abramowitz M, et al "Serum phosphate levels and mortality in patients with normal renal function attending inner-city outpatient clinics" NKF Meeting 2008; Abstract 39.