#2509 Longitudinal haemoglobin levels and mortality in an elderly population with advanced chronic kidney disease: insights from the EQUAL Study

Abstract Background and Aims The relationship between haemoglobin (Hb) levels and clinical outcomes in chronic kidney disease (CKD) is more intricate than previously understood, as both low and high values can have detrimental effects. Although the association between Hb and patient outcomes is widely reported, few studies have focused on older patients with advanced CKD or have explored this relationship both before and after kidney replacement therapy (KRT) initiation. The primary objective of this study is to describe the relationship between Hb levels and mortality in older patients with advanced CKD in the continuum from pre-KRT phase to subsequent KRT initiation, and to evaluate any effect modification by KRT initiation. Method We used data from the European QUALity Study on treatment in advanced CKD (EQUAL) cohort which includes patients aged ≥65 years and eGFR ≤20 ml/min per 1.73 m2 from six European countries. The association between Hb and all-cause and cause-specific mortality (cardiovascular [CV] and non-cardiovascular [non-CV] mortality) was explored in three separate CKD populations: pre-KRT, KRT and transition-CKD (transition phase surrounding the start of KRT) population. Multivariable adjusted time-dependent Cox regression models and natural cubic splines were used to investigate the association between longitudinal Hb levels (as categorical and continuous variable) and mortality risk in the pre-KRT and KRT population. The area under the Hb trajectory (AUC Hb) before KRT initiation was calculated to evaluate the relationship between the cumulative exposure of Hb before KRT initiation and all-cause mortality after KRT initiation in the transition-CKD population. We evaluated effect modification by pre-specified variables on the relationship between Hb and mortality. Results Overall, 1716 CKD patients were included in the study. During a median follow up of 3.0 years (IQR 2.6), 614 patients died, resulting in a 5-year survival probability of 51% (95% CI 48, 55). Low Hb levels were consistently associated with increased mortality risk in all study populations (Figure). A U-shaped association between both the lowest and highest levels of Hb and mortality risk was observed (Figure). A similar association was observed for non-CV mortality, with lower Hb values associated with an increased mortality risk (Figure). Conversely, a trend towards higher risk of cardiovascular death was observed as Hb levels increase, especially in the whole population and pre-KRT population (Figure), albeit not statistically significant. Low cumulative exposure of Hb in the transition-CKD population was associated with increased mortality risk after transitioning to KRT (p = 0.047). In subgroup analysis diabetic patients were more susceptible to low Hb levels (p-value for interaction = 0.003) while patients on ESA therapy (p-value for interaction = 0.005) or initiating KRT (p-value for interaction = 0.024) showed increased mortality risk for high Hb levels. Conclusion The relationship between Hb levels and mortality in advanced CKD patients is complex. Low Hb levels consistently indicate an increased risk of mortality, while high Hb levels also appear to have a negative effect, increasing mortality risk, particularly in patients receiving ESA therapy and after KRT initiation. Our work emphasizes how different stages of advanced kidney disease (i.e., before and after initiating KRT), as well as comorbidities such as diabetes or medications such as ESA therapy can significantly influence the relationship between Hb and mortality.