Albuminuria Is Associated with Endothelial Dysfunction in Sickle Cell Disease

Introduction: Albuminuria is a common and sensitive marker of glomerular injury in SCD. We and others have previously observed the association of albuminuria with echocardiographically-derived tricuspid regurgitant jet velocity, systolic blood pressure, and history of stroke, suggesting a shared vasculopathic pathophysiology. However, the pathogenesis of albuminuria in SCD remains poorly defined. The purpose of this study was to evaluate the association of albuminuria with endothelial dysfunction, assessed non-invasively using ultrasound imaging of the brachial artery in SCD patients. In addition, we evaluated the association of albuminuria assessed by spot urine collection with 24-hour urine collections for protein.

Methods: Spot urine measurements for microalbumin/creatinine ratio (UACR) were obtained during the "non-crisis," steady state over 2-3 visits, with a 24-hour urine collection for assessment of protein at the final visit. We obtained routine laboratory tests, including markers of hemolysis (lactate dehydrogenase, and total and indirect bilirubin) at the baseline visit. Endothelial function was assessed non-invasively at the final visit using a brachial artery shear stress method, with measurements of flow-mediated dilation (FMD), an index of endothelium-dependent vasodilation and nitroglycerin-mediated dilation (NTMD), a measure of endothelium-independent vasodilation. Patients on renin-angiotensin-aldosterone system (RAAS) blocking agents (ACE-inhibitors, angiotensin receptor blockers, direct renin inhibitors and mineralocorticoid receptor blockers), as well as those on hydroxyurea were on a stable dose for at least 3 months prior to enrollment. We evaluated associations between measures of endothelial function and other study covariates using Spearman's correlation coefficient for continuous covariates, and the Kruskal Wallis test for categorical covariates. Linear regression analysis was used to adjust for baseline arterial diameter. Reported p-values are for individual tests, unadjusted for multiple comparisons.

Results: Spot urine samples were obtained in 23 subjects with sickle cell anemia (female = 13, male = 10), mean age, 43.3 ± 12.0 years, with normoalbuminuria (UACR < 30 mg/g creatinine) in 7 subjects, microalbuminuria (UACR = 30 - 299mg/g creatinine) in 9 subjects and macroalbuminuria (UACR ³300 mg/g creatinine) in 7 subjects. Albuminuria (assessed by spot microalbumin/creatinine ratio) was significantly correlated with FMD (ρ = -0.45, p = 0.031). However, albuminuria was no longer significantly associated with FMD when controlled for baseline arterial diameter (p = 0.09). Although FMD appeared to be lowest in patients with macroalbuminuria, no significant difference was observed when FMD was evaluated in the 3 albuminuria categories (p = 0.16). No significant correlations were observed between NTMD and albuminuria (ρ = 0.35, p = 0.20). Surprisingly, no significant correlations were observed between FMD and hemoglobin, fetal hemoglobin, lactate dehydrogenase or indirect bilirubin. We observed a strong correlation between albuminuria assessed by spot microalbumin-creatinine ratio and total urinary protein assessed by 24-hour urine collection (ρ = 0.90, p < 0.0001). Furthermore, albuminuria assessed by spot microalbumin-creatinine ratio was strongly correlated with urine aliquots for microalbumin-creatinine ratio obtained from the 24-hour urine collection (ρ = 0.97, p < 0.0001).

Conclusion: In this exploratory study, we found that albuminuria is significantly correlated with FMD in patients with SCD, suggesting that endothelial dysfunction may contribute to the pathogenesis of this complication in SCD. We also find a very strong correlation between spot urine assessments for microalbumin/creatinine ratio and 24-hour urine collections for protein, providing validation for the use of spot urine assessments for the evaluation of albuminuria in patients with SCD.