Recent studies have indicated that higher target hemoglobin (Hb) levels are associated with an increased risk of mortality in patients with chronic kidney disease. Use of high erythropoietin (EPO) doses has been suggested as one factor that might be involved in these observations. A variety of factors appear to influence EPO sensitivity and responsiveness, including iron sufficiency, blood loss, adequacy of dialysis, infection, and inflammatory disorders. Several markers of inflammation have been associated with EPO responsiveness. By analyzing these markers, clinicians may be able to make predictions concerning EPO sensitivity or responsiveness in hemodialysis patients. Recently, a study hypothesized that the changes in white blood cell (WBC) count and serum albumin levels—along with other experimental biomarkers such as monocyte chemoattractant protein 1 (MCP1) and malondialdehyde (MDA)—could predict EPO sensitivity.

In the study, several potential markers of ESA sensitivity – WBC count, serum albumin concentration, levels of MDA and MCP1, and tests of iron status and their changes over time, as well as ESA dose and iron therapy – were recorded over a 3-month period in a population of 82 prevalent hemodialysis patients (mean age 53.3 years). Using a series of covariance models with various fixed parameters, the results showed that patients with higher Hb levels also had higher levels of serum albumin and vice versa. Serum albumin was also a significant predictor of baseline Hb and EPO sensitivity. However, no independent relationships were found between WBC count, MCP1, or MDA levels and EPO sensitivity. Tests of iron sufficiency and use of intravenous iron therapy were also not predictive of EPO responsiveness.

The significant relationship between serum albumin levels and EPO sensitivity is an important finding because albumin is an excellent marker of overall health for patients undergoing dialysis. Observations of increased serum albumin concentrations may indicate improving infection, inflammation, oxidative stress, and nutrition. Clinicians may be able utilize serum albumin as a routine, bedside tool to determine if decreases in Hb are caused by infection, inflammation, or other factors. Future clinical trials on EPO responsiveness with stratification of patient populations according to this biomarker may allow for a better understanding of modifiable risk factors that arise during hemodialysis.

Serum albumin is strongly associated with erythropoietin sensitivity in hemodialysis patients. Agarwal R, Davis JL, Smith L. Clin J Am Soc Nephrol. 2008 Jan;3(1):98-104.

NAAC Expert Commentary:
Much recent research and commentary has focused on outcomes associated with specific Hb target levels (1,2), causes and consequences of Hb variability (3,4), and the relationships between markers of inflammation or nutrition and responsiveness to erythropoiesis stimulating agents (ESAs) and other clinical outcomes (5,6) in patients with CKD. Retrospective, observational cohort studies derived from large administrative data bases, showed us that higher achieved Hb levels in hemodialysis patients were associated with better outcomes (i.e. mortality and hospitalizations) and lower Hb levels with poorer outcomes (7,8). We now have evidence from prospective clinical trials in hemodialysis patients (9) and patients with CKD who are not on dialysis (1,2) that targeting higher Hb levels does not result in better outcomes, and in fact may cause higher mortality risk. The reasons for this are not yet entirely clear. What is emerging is a picture of complex and as of yet still poorly understood interactions between achieved Hb level, nutritional status, inflammatory responses, ESA responsiveness, ESA doses used to achieve (or attempt to achieve) targeted Hb levels, Hb variability, and iron therapies that are intertwined with individual patient characteristics and impact clinical outcomes. It does seem that using ESAs with the intent of achieving the same Hb level in all CKD and dialysis patients is neither possible nor advisable.

In this study by Agarwal et al, all of the patients received erythropoietin (EPO) subcutaneously. There was a relatively high proportion of patients dialyzed with catheters (36%) while use of fistulas was infrequent (27%). Although the authors suggest that patients with extremes of Hb levels tended to have greater Hb variability than those in the middle, this should be treated as a preliminary observation requiring further study and if real, is likely due mostly to efforts to raise the Hb in those with low baseline levels and reduce the Hb in those with high baseline levels. In modeling the relationship between all study parameters and changes in Hb during the 3 months of the study, baseline EPO dose, EPO dose over time, baseline serum albumin, serum albumin over time, and baseline WBC count were predictors of Hb change. After accounting for EPO dose, only serum albumin was related to baseline Hb and predictive of the 3-month change in Hb. As has become reported previously, serum ferritin and transferrin levels were poor predictors of Hg changes (TSAT was not studied).

As the authors suggest, these results might be relevant to patient care because changes in serum albumin may help anticipate changes in ESA responsiveness and provide clues as to the cause of changes in Hb levels. An increasing serum albumin might be used to anticipate an increase in ESA responsiveness and trigger a reduction in ESA dose. A fall in Hb associated with a low or declining albumin suggests an inflammatory and/or infectious process is present, while a fall in Hb with a normal or increasing albumin would be more suggestive of blood loss or iron insufficiency. Recognizing that clinical outcomes in individual patients are perhaps related more to patient-specific factors that influence changes in Hb, serum albumin, and ESA responsiveness than the absolute Hb level itself may help us get out of the trap of mindlessly increasing the ESA dose in hyporesponsive patients in pursuit of a specific Hb level and allow us to focus more on what is actually happening with our patients.

References

1. Drueke TB, et al. N Engl J Med 2006;355:2071-84.
2. Singh AK, et al. N Engl J Med 2006;355:2085-98.
3. Berns JS, et al. Kidney Int 2003;64:1514-21.
4. Brunelli SM, et al Clin J Am Soc Nephrol 2008;3:777-82.
5. Locatelli F, et al. Nephrol Dial Transplant 2006;21: 991-8.
6. Kalantar-Zadeh K, et al. Am J Kidney Dis 42:761-73.
7. Xia A, et al. J Am Soc Nephrol 1999;10:1309-16.
8. Collins AJ, et al. J Am Soc Nephrol 2001;12:2465-73.
9. Besarab A, et al. New Engl J Med 1998;339:584-90.

Last Modified: August 14, 2008