Bicarbonate is the principal buffer in dialysate, but only few studies have assessed outcomes of patients treated with different dialysate bicarbonate (DB) concentrations. A recent study in AJKD looked at the association of DB concentration with mortality in the Dialysis Outcomes and Practice Patterns Study (DOPPS), and found that high DB may be associated with increased morbidity and mortality. Dr. Francesca Tentori (FT), lead author of the article, discussed the findings of the study with Dr. Abdo Asmar (eAJKD), eAJKD contributor.

eAJKD: Why are you interested in this topic?

FT: The DOPPS is a longitudinal prospective cohort study of hemodialysis practices in more than 20 countries. The DOPPS focuses on modifiable practices as a means to improve patient outcomes. Many factors, such as patient age and diabetes status, are strong predictors of mortality, but are not modifiable. Other factors of dialysis treatment, such as the dialysate composition, can be adjusted and may have an effect on patient outcomes. Using DOPPS data, we have the ability to observe variability in practices between countries. Given that sodium bicarbonate is the principal buffer used in dialysate and that the optimal concentration is unknown, we wanted to study the DB prescription and its association with clinical outcomes. As we were conducting these analyses, dialysate alkali concentration became the subject of a communication issued by the FDA, and interest in this topic has grown within the dialysis community.

eAJKD: Could you discuss DB concentration versus total base levels? Would the study results have changed if you had looked at total buffer and not just DB?

FT: My sense is that the majority of nephrologists may not be aware of this difference. The DOPPS has asked about DB for a number of years. We realized that we might be underestimating the actual total base in the dialysate. We added a specific question about total base instead of only DB. Interestingly, similar numbers were reported back. It appears that the difference between the two was not evident to many people who responded. Providers routinely prescribe DB forgetting that the acetate (about 2-8 mEq/L) component of the bath, which gets metabolized in the liver to bicarbonate. I think we can say that the association between the higher DB and mortality might be greater than we think, given that we are underestimating the total base by just being reporting the DB.

eAJKD: Were you surprised to find the increased mortality was predominantly driven by infection-related complications rather than cardiovascular causes? How do you explain this?

FT: I was certainly surprised by this finding. We tested the association with infections, postulating that it would be null. Our thoughts going into this were that higher DB would lead to the development of intradialytic and postdialytic alkalosis that might act two ways. First, during the treatment, patients might have a rapid transcellular shift of electrolytes resulting in higher incidence of arrhythmias and sudden death. Second, during the interdialytic period, alkalosis might increase precipitation of calcium leading to cardiovascular calcifications and perhaps more cardiovascular events. We found that there is a very strong association between higher DB and mortality related to sepsis. The type of infection and the reason for infection is less clear. There is some evidence from basic science studies that extracellular acidosis may be associated with an improved response of the dendritic cell antigen-presenting capacity. One could speculate that alkalosis may lead to a decreased immune response and more infections, but that is speculative at this time, and more studies are needed to answer this question.

eAJKD: Your study found that tailoring the DB concentration to a specific acid-base status did not result in any survival benefit. How do you explain this finding? What should clinicians use to determine the optimal DB concentration?

FT: Even though we have applied a robust analysis, the design of observational studies does not allow us to prove a causal relationship. Hence, we cannot state that higher DB leads to a higher mortality, or that there is no benefit in tailoring the dialysate prescription to a specific patient’s acid base status. Many centers, including about 50% of centers in North America, did not tailor the prescription of the DB to a specific patient’s acid base status, which could have limited our ability to detect potential benefits. Therefore, the lack of association with improved outcomes should not prevent clinicians from making a tailored prescription of DB for a given patient. The patient’s specific clinical status should always be the main factor in deciding the dialysis prescription.

To view the article abstract or full-text (subscription required), please visit AJKD.org.