Hemoglobin (Hb) variability, which is the oscillation or fluctuation of an individual’s Hb levels over time, has been associated with adverse outcomes in anemic patients and was recently identified as a potential therapeutic target. Currently, clinical research has not determined a clear method to define or measure Hb variability and its relationship to adverse outcomes, nor have studies assessed this relationship outside of dialysis patient populations. Therefore, the study by Boudville et al examined the relationship of Hb variability and mortality in nondialysis chronic kidney disease (CKD) patients receiving erythropoiesis-stimulating agents (ESAs), and aimed to identify specific factors that are associated with Hb variability.

The study included 6,165 nondialysis CKD patients, half of whom were receiving ESA treatment. These patients were enrolled at various healthcare centers from seven countries between January 2003 and October 2005, and received at least three Hb measurements within a 6-month base-line period during enrollment. To consistently evaluate and establish a definition of Hb variability, a number of study parameters were used including the mean change and standard deviation of the fluctuation in Hb between measurements, Hb amplitude between the highest and lowest measurement, and a classification system of patient Hb levels.

Using these parameters, as well as descriptive statistics that adjusted for factors such as age and gender, the study revealed that Hb variability is increased in nondialysis CKD patients and is associated with an increased risk of death. In addition, patients using ESAs showed a higher rate of Hb variability compared to those not using ESAs. Furthermore, a significant number of patients (31.7%) who used ESAs throughout the entire base-line period died by the completion of follow-up, consistent with previous studies.

Although patients in the ESA group showed higher risk for mortality, patients not using ESAs also showed a significant risk for death with increased Hb variability. This finding raises many questions about whether Hb variability can causally affect mortality or if it is simply an epiphenomenon. For example, age, male gender, and presence of diabetes were found to be the most significant predictors of mortality in the ESA group, suggesting that multiple factors, rather than ESA treatment by itself, are involved in Hb variability and mortality. However, because the ESA group showed higher Hb variability and greater risks for mortality, the authors concluded that researchers should consider ESA treatments as a focal point for future studies so that Hb variability can be better understood as a marker of illness and comorbidity.

Boudville NC, Djurdjev O, Macdougall IC, de Francisco AL, Deray G, Besarab A, Stevens PE, Walker RG, Ureña P, Iñigo P, Minutolo R, Haviv YS, Yeates K, Agüera ML, Macrae JM, Levin A. Hemoglobin Variability in Nondialysis Chronic Kidney Disease: Examining the Association with Mortality. Clin J Am Soc Nephrol. 2009 May 7.

NAAC Expert Commentary:
Variability in Hb levels from day to day and even hour to hour has been documented in normal individuals and likely reflects non-constant erythropoietin (EPO) output, bone marrow response, and fluid status. However, there has been little work done on the clinical implications of such variability in normal patients. In patients with chronic kidney disease (CKD) on dialysis, Hb variability has been described in many studies, work driven primarily by reimbursement policies for recombinant erythropoietin that set limits on Hb values that are acceptable to allow for payment for ESAs. Such studies have demonstrated substantial Hb variation and attempts have been made to correlate such variation with clinical outcomes, with conflicting results. The myriad factors that may contribute to Hb variability in this complex, ill population is the likely explanation for the difficultly in demonstrating cause and effect. The current study attempts to look at a less complex population, those with CKD not yet on dialysis, including patients not yet receiving ESAs. The results suggest an association between Hb variability and adverse clinical outcomes, with a greater effect in patients receiving ESAs. The study design precludes cause and effect conclusions, but is hypothesis generating – that is, it is plausible that Hb variability is itself a contributor to poor outcomes. An equally plausible hypothesis is that Hb variability is a marker of severity of illness – sicker patients have more variability and therefore poorer outcomes. Only carefully designed controlled trials will provide answers that are more definitive.

Last Updated: June 16, 2009