Sex, Age, and the Association of Serum Phosphorus With All-Cause Mortality in Adults With Normal Kidney Function.

BACKGROUND

High serum phosphorus levels are associated with cardiovascular morbidity and mortality in kidney disease. Although serum phosphorus levels possibly influence on mortality in individuals without kidney disease, this is uncertain because of the variable sex- and age-based distribution of serum phosphorus levels.

STUDY DESIGN

Observational cohort study.

SETTING & PARTICIPANTS: Clinical and biochemical data were collected from 138,735 adults undergoing routine health checkups in 3 tertiary hospitals. Individuals with estimated glomerular filtration rates < 60 mL/min/1.73 m2 and urine dipstick albumin ≥ 1+ were excluded.

PREDICTOR

Sex-specific quartiles of serum phosphorus and sex.

OUTCOMES

All-cause mortality.

RESULTS

The study included 92,756 individuals. Generally, women showed higher serum phosphorus levels than men. In women, serum phosphorus levels increased with age until 60 years old, then decreased with age. Men with higher serum phosphorus levels were younger and less likely to have hypertension, whereas women with higher serum phosphorus levels were older and more likely to have diabetes and hypertension. During a median follow-up of 75 months, 1,646 participants died. In the overall population, higher serum phosphorus levels were an independent predictor for all-cause mortality after adjustment (adjusted HR for the highest vs. lowest quartile, 1.34; 95% CI, 1.15-1.56; P<0.001). We observed that this increased risk was present in men but not in women (adjusted HR of 1.43 [95% CI, 1.22-1.68] vs. 1.01 [95% CI, 0.76-1.33]), but interaction by sex was not significant (P=0.8).

LIMITATIONS

A single phosphorus measurement and low power to test for interactions by sex and age.

CONCLUSIONS

We demonstrated that higher serum phosphorus levels influenced all-cause mortality in individuals with normal kidney function. Our findings suggest that the association may differ by sex, but future studies with adequate power to test for effect modification are needed to confirm our findings.