The uremic syndrome is attributed to progressive retention of compounds that, under normal conditions, are excreted by the healthy kidneys. p-cresol sulfate (PCS), a prototype protein-bound uremic retention solute, has been shown to exert toxic effects in vitro. Recent studies have identified relations between increased levels of PCS and indoxyl sulfate (IS) and adverse clinical outcomes in hemodialysis patients. We explored the relationship between free and total PCS and IS with infection-related hospitalizations (IH) and septicemia in 2 cohorts, Choices for Healthy Outcomes in Caring for end-stage renal disease (ESRD) Study (CHOICE) and Hemodialysis Study (HEMO).
We measured free and total levels of PCS and IS in stored specimens in CHOICE, a cohort of 464 incident hemodialysis patients enrolled in 1995 to 1998 and followed for an average of 3.4 years and in a prevalent dialysis cohort of 495 patients enrolled in HEMO from 1995 to 2000 and followed for an average of 4.4 years. We measured free PCS and IS using mass spectroscopy. The 2 cohorts were linked to United States Renal Data System (USRDS) Medicare billing records to ascertain IH over follow-up. We examined the association of free and total levels of PCS and IS with IH and septicemia using multilevel Poisson regression models adjusted for demographics, comorbidities, clinical factors, and laboratory tests including residual kidney function. We stratified patients a priori based on gastrointestinal (GI) disease as PCS and IS are produced in colon.
In CHOICE, highest tertile of free PCS in multivariable model was associated with 50% higher risk of IH [95% CI = 1.01–2.23] compared with lowest tertile in patients with no-GI disease. A significant trend was noted between greater levels of free PCS and septicemia in no-GI disease group in both cohorts, while no association was noted in GI disease group. Total PCS concentrations were not associated with either IH or septicemia in either cohort. No significant risk of IH or septicemia was noted with higher levels of free or total IS in either GI or no-GI disease group.
These results suggest an association between higher concentrations of free PCS and infection-related and sepsis-related hospitalizations in hemodialysis patients. Better methods of dialysis should be developed to evaluate the utility of removing PCS and its effect on the outcome and also therapies to decrease gastrointestinal tract production of uremic solutes.