Acute hospitalization prior to dialysis predicts post-dialysis harm, UC research shows

Hospitalizations in the pre-dialysis period adversely impact dialysis outcomes according to new research from the University of Cincinnati (UC). The study, published online in PLOS ONE finds that pre-dialysis hospitalization is an independent predictor of one-year mortality in dialysis patients.

Each year in the United States, over a half million patients receive dialysis for chronic kidney disease and approximately 100,000 new patients start dialysis. Dialysis patients experience high morbidity and an average mortality of 20 percent at one year following dialysis initiation. In addition, dialysis care is expensive—Medicare spends an average of $89,000 per patient per year for total care of a dialysis patient.

“Given that patients with chronic kidney disease, a precursor to end stage kidney disease, face an increasing burden of hospitalizations, our study focused on the impact of pre-dialysis acute care hospitalization on clinical outcomes in patients with ESKD [end-stage kidney disease],” says Silvi Shah, MD, assistant professor, Division of Nephrology, Kidney CARE Program at the UC College of Medicine, and lead author of the study.

The study uses data from the United States Renal Data system with linked data from Medicare parts A and B to examine a cohort of 170,897 adult patients who initiated dialysis. The analysis shows that three-fourths of patients with incident ESKD had at least one hospitalization event during the two-year pre-ESKD period.

“The study found that both the cause and frequency of acute care hospitalizations during the pre-ESKD period significantly increased mortality at one year after the start of dialysis,” says Shah. “For instance, a cardiovascular related pre-dialysis hospitalization was associated with a 63 percent increase in one-year mortality. The impact is much more pronounced in those with a history of both cardiovascular and infection-related pre-dialysis hospitalization, which increased the risk of death by 91 percent.”

Additionally, the study shows that the adjusted odds of hemodialysis vs. peritoneal dialysis as the initial dialysis method were higher, whereas adjusted odds to initiate hemodialysis with an arteriovenous access vs. central venous catheter were lower in patients with any type of pre-dialysis hospitalization.

“Pre-ESKD health is a critical determinant of post-dialysis outcomes,” says Charuhas Thakar, MD, professor and division director of nephrology at the UC College of Medicine, and a co-author of this study. “By leveraging the unique strengths of two large national registries, we hypothesized that hospitalization during the pre-ESKD period, as an indicator of health status, will predict one-year dialysis outcomes. Our findings confirm this notion, and suggest that more attention and resources need to be directed in improving coordinated care during pre-dialysis health status to meaningfully impact both method of dialysis as well as one year outcomes.”

“When a patient initiates dialysis, we seldom recognize the impact of acute care hospitalizations that may have occurred in the preceding two-year period, and sometimes clinicians don’t even have access to this information due to complex care transitions,” Shah says. “Reducing these hospitalizations could require multi-level changes to chronic kidney disease care. Since dialysis initiation is a life-changing event, this transition of care may need to be customized based on pre-dialysis acute care events.”

In addition to Thakar, co-authors in the study include Anthony Leonard, PhD, research associate professor in the Department of Family and Community Medicine, Karthikeyan Meganathan and Annette Christianson, research associates in the Department of Biomedical Informatics, all in the UC College of Medicine.

Related Stories

3

Pharmacy Times: How AI, machine learning can benefit...

March 25, 2024

Pharmacy Times highlighted a presentation from the University of Cincinnati's Shawn Xiong discussing the potential for artificial intelligence and machine learning to bolster pharmaceutical development and research.

Debug Query for this