Renal Disease in Sickle Cell: Clinically Varied and Associated with Increased Mortality

Abstract 90

The Walk-Phasst Study of sickle cell disease (SCD) affords a unique opportunity to examine renal function in a large number of adults with SCD. Extensive clinical and laboratory data were obtained on 463 adults with HbSS and 127 adults with HbSC. Where possible, correlates for estimated glomerular filtration rate (eGFR, calculated by the 4-value MDRD equation) and albuminuria/proteinuria were also evaluated in historical data, from SS adults in the CSCCD cohort and the Multicenter Study of Hydroxyurea (MSH) in sickle cell disease.

Adjusted decline in eGFR was more rapid in SS compared with SC, at −2.6 and −0.92 ml/min/1.73 m2/year, respectively (p<0.001). In SS, similar declines in cross-sectional eGFR with age were seen in adults in the CSCCD (n=705) and MSH (n=298) cohorts, at −2.9 and −1.9 ml/min/year (BSA corrected), respectively. Multivariate analysis of the SS cohort in of Walk-Phasst revealed that depressed eGFR values were associated with an elevated estimated pulmonary arterial systolic pressure (as reflected in a higher tricuspid regurgitant jet velocity, measured by echocardiogram, p=0.007) and with evidence for inflammation (an elevated white blood cell count, p=0.018). In SC adults, in contrast, lower eGFRs were primarily associated with a diminished erythroid reserve (depressed absolute reticulocyte counts, P=0.005).

Albuminuria (≥30 mg albumin/gm creatinine) was detectable in 66 and 41 % of adults with HbSS and HbSC in Walk-Phasst (185/287 evaluable SS and 25/61 evaluable SC, p=0.001). In Walk-Phasst, albuminuria in SS was associated with evidence of increased red cell destruction (lower total hemoglobin (P=0.07), higher LDH (P<0.001), and higher reticulocyte count (P=0.017)). In SC, albuminuria was associated with a higher LDH (P=0.01). 159/657 (24%) of adult SS subjects in the CSCCD cohort had trace to 4+ proteinuria, using a method (urine dipstick analysis) that is less sensitive and less quantitative than the albumin-to-creatinine ratio used in Walk-Phasst. Multivariate analyses again suggested a strong association between a depressed Hgb and proteinuria (P<0.001) in CSCCD, and LDH was also associated with proteinuria, in univariate analyses (P<0.001). In Walk-Phasst the absence of albuminuria in all subjects with SCD was associated with lesser mortality in multivariate analyses (Hazard ratio 0.62 (0.4–0.9, P=0.02). No similar association was seen between eGFR and mortality.

Subjects with SS in Walk-Phasst had depressed serum bicarbonate levels, of 23.8±3.4 compared with 24.7±3.2mEq/dL in SC (p<0.005) and 24.8±3.3 mEq/dL in the non-CKD general population (Shah et al, 2009). In multivariate analyses, acidemia was associated with both increased destruction and decreased production of red cells, e.g. higher EPO (P=0.003) and total bilirubin levels (P<0.001), but lower reticulocyte counts (P=0.06). Impaired physiologic functioning in acidemic subjects with HbSS, manifest as a depressed 6MWD (P<0.001) and a lower eGFR (P<0.001), was seen. No effect of bicarbonate level on mortality in SCD patients was evident in Walk-Phasst.

In conclusion, renal function is perturbed in sickle cell syndromes in several ways, including more rapid decrement in eGFR, highly prevalent albuminuria, and, in SS, marked abnormalities in acid-base balance. Albuminuria is associated with anemia in two large cohorts of sickle cell disease patients and, in Walk-Phasst, with evidence for enhanced red cell destruction. Importantly, albuminuria correlated with an increased risk for mortality in sickle cell disease.

(We acknowledge the many contributions made by the Walk-PHASST Investigators: Badesch DB, Barst RJ, Castro OL, Girgis RE, Gibbs JS, Goldsmith JC, Hannoush H, Hassell KL, Kato GJ, Krishnamurti L, Lanzkron S, Machado RF, Morris CR, Novelli EM, Rosenzweig EB, Sachdev V, Schraufnagel DE, Taylor JG 6th.)