Serum phosphorus levels and cognitive performance in the Framingham Offspring cohort Study
Daniluk, Daniel Alexander
MetadataShow full item record
BACKGROUND: With the proportion of the world’s elderly population continuing to increase dramatically, tremendous amounts of research have focused on detecting the earliest signs of cognitive impairment before the onset of dementia. The pathophysiology of dementia is complex and recent genetic and biomarker studies have identified new biological pathways that might modify the risk of dementia. One potential risk factor, altered serum phosphorus levels, has been studied with respect to its potential impact on human cognition. The association between serum phosphorus and cognition needs further investigation using a population that is free of CKD. OBJECTIVE: I used data from the Framingham Offspring Study (FOS) cohort to investigate the cross-sectional association between measured serum phosphorus within the normal range and cognitive performance in women and men. METHODS: Participants from the FOS who attended the ninth examination cycle were included in this analysis (N=1253). The Wechsler Memory Scale: Logical Memory – Immediate and Delayed recall (LM-IR and LM-DR) and Visual reproduction – Immediate and Delayed recall (VR-IR and VR-DR) tests were used to assess verbal and visual memory, respectively. Times for the Trailmaking Test – Parts A and B along with the difference between the two tests (B-A) were used to assess attention, psychomotor speed, and executive functioning. Participants were categorized according to levels of fasting serum phosphorus as follows: Low Phosphorus - 2.0-<3.1 mg/dL for men and 2.6-<3.6 mg/dL for women, Moderate Phosphorus - 3.1-<3.6 mg/dL for men and 3.6-<3.9 mg/dL for women, High Phosphorus - 3.6-<5.2 mg/dL for men and 3.9-<5.3 mg/dL for women. Mean cognitive scores were compared across categories using a least squares general fit linear model. Multivariable logistic regression models were used to investigate the association between higher serum phosphorus levels and odds of cognitive impairment within each cognitive test. RESULTS: We did not find any statistically significant differences in mean scores on the Logical Memory, Visual Reproduction, and Trailmaking – Part A tests among the categories of fasting serum phosphorus. There was no association between higher serum phosphorus levels and odds of cognitive impairment on any of the verbal and visual memory tests. In men, higher serum phosphorus levels were associated with poorer performance on the Trailmaking Test – Part B (High phosphorus: -0.50 ± 0.04, Moderate phosphorus: -0.40 ± 0.03, Low phosphorus : -0.33 ± 0.04; P-trend: <0.002) and the difference in log-transformed times between the Trailmaking Test – Part B and A tests (High phosphorus: -1.16 ± 0.02, Moderate phosphorus: -1.10 ± 0.02, Low phosphorus: -1.07 ± 0.02; P-trend: <0.004). Higher serum phosphorus levels were associated with an 80% greater odds of having a cognitively impaired score on Trailmaking Test – Part B-A in men (OR: 1.81, 1.11-2.94), and this association was strengthened when adjusting for additional confounding variables (OR: 2.02, 1.15-3.54). There was no such association in women. Using a cubic spline regression analysis, we treated serum phosphorus as a continuous variable and observed a positive linear association between phosphorus and total time or Trailmaking – Part B-A in men. In particular, the odds of cognitive impairment increased at levels of phosphorus above 3.5 mg/dL. CONCLUSIONS: This study demonstrated that higher serum phosphorus levels were associated with poorer performance on the Trailmaking Test – Part B, and in times represented by the Trailmaking Test – Part B-A time in men. We also observed that higher fasting serum phosphorus levels as a continuous variable were associated with increased odds of cognitive impairment on Trailmaking Test – Part B-A in men. We found no association between higher serum phosphorus levels and lower verbal and visual memory scores or increased odds of cognitive impairment on those scores. Since fewer women had cognitive impairment on these test, statistical power was limited for some of these analyses. Future studies are necessary to examine the mechanistic pathways by which serum phosphorus could impact cognition and whether these effects are independent of cardiovascular disease.