Abstract
Background and objectives
Sickle cell disease (SCD) is an inherited anemia that afflicts millions worldwide. Kidney disease is a major contributor to its morbidity and mortality. We examined contemporary and historical SCD populations to understand how renal disease behaved in hemoglobin SS (HbSS) compared with HbSC.
Design, setting, participants, & measurements
Kidney function was examined in the multicentered Treatment of Pulmonary Hypertension and Sickle Cell Disease with Sildenafil Therapy (Walk-PHaSST) Trial (HbSS=463; HbSC=127; years 2007–2009) and historical comparator populations from the Cooperative Study of Sickle Cell Disease (CSSCD; HbSS=708) and the Multicenter Study of Hydroxyurea in Sickle Cell Disease (MSH; HbSS=299).
Results
In adults with SCD, eGFR was lower among older individuals: −1.78 ml/min per 1.73 m2 per year of age (95% confidence interval [95% CI], −2.06 to −1.50; Walk-PHaSST Trial), −1.75 ml/min per 1.73 m2 per year of age (95% CI, −2.05 to −1.44; MSH), and −1.69 ml/min per 1.73 m2 per year of age (95% CI, −2.00 to −1.38; CSSCD) in HbSS compared with −1.09 ml/min per 1.73 m2 per year of age (95% CI, −1.39 to −0.75) in HbSC (Walk-PHaSST Trial). Macroalbuminuria was seen in 20% of participants with SCD (HbSS or HbSC; P=0.45; Walk-PHaSST Trial), but microalbuminuria was more prevalent in HbSS (44% versus 23% in HbSC; P<0.002). In the Walk-PHaSST Trial, albuminuria was associated with hemolysis (higher lactate dehydrogenase, P<0.001; higher absolute reticulocyte count, P<0.02; and lower Hb, P=0.07) and elevated systolic BP (P<0.001) in HbSS. One half of all participants with HbSS (20 of 39) versus one fifth without (41 of 228) elevated tricuspid regurgitant jet velocity (≥3 m/s; adverse prognostic indicator in SCD) had macroalbuminuria (P<0.001). In the CSSCD, overt proteinuria, detected (less sensitively) by urine dipstick, associated with higher 3-year mortality (odds ratio, 2.48; 95% CI, 1.07 to 5.77). Serum bicarbonate was lower in HbSS (23.8 versus 24.8 mEq/dl in HbSC; P<0.05) and associated with reticulocytopenic anemia and decreased renal function.
Conclusions
In SCD, albuminuria or proteinuria was highly prevalent, in HbSS more than in HbSC. Proteinuria associated with mortality in HbSS. Older individuals had a lower than expected eGFR, and this was more prominent in HbSS. Current management does not routinely address renal complications in SCD, which could plausibly reduce morbidity and mortality.
Keywords: acidosis; albuminuria; chronic kidney disease; anemia, sickle cell; blood pressure; glomerular filtration rate; humans; hydroxyurea; hypertension, pulmonary
Introduction
Sickle cell disease (SCD) is an inherited anemia resulting from mutations in the β-globin chain of adult hemoglobin (HbA; α2β2). Homozygous HbSS (α2βS2) is most common, but approximately one third of adults with SCD have compound heterozygous HbSC (α2βsβc). Anemia, painful vaso–occlusive episodes, hemolysis, widespread vasculopathy, and early mortality are characteristic of HbSS (1–4). Retinopathy, avascular necrosis of the bones, and chronic pain are more typical of HbSC. Participants with HbSC have a more normal life expectancy (5,6).
Kidney disease is seen in most adults with SCD (7–10) and may affect glomerular and/or tubular function (11,12). CKD in SCD is associated with excess mortality at clinical baseline (4) and during treatment with hydroxyurea (the current standard of care for HbSS [13]). However, the clinical picture is incomplete. Reports of CKD in SCD have focused primarily on patients with HbSS (8,14,15). Furthermore, the life expectancy of adults with SCD has lengthened, and more recent multicentered studies are likely to be more informative about the current status of renal disease in this population.
The Treatment of Pulmonary Hypertension and Sickle Cell Disease with Sildenafil Therapy (Walk-PHaSST) Trial was a large, modern, and well phenotyped study of people with homozygous or compound heterozygous SCD. It assembled clinical and laboratory data on 720 participants recruited between 2007 and 2009 and afforded a unique opportunity to evaluate and contrast renal disease in people with HbSC or HbSS relative to cardiovascular phenotype, erythroid reserve, hemolysis, and mortality. The objective of this study was to evaluate renal function and acid-base complications in HbSS and HbSC as they related to erythroid reserve, hemolysis, and mortality in adults (≥18 years of age) from the Walk-PHaSST Trial population.
Conclusions about renal disease in HbSS from the Walk-PHaSST Trial were tested for validity and corroboration in prior observational (the Cooperative Study of Sickle Cell Disease [CSSCD]) and treatment (the Multicenter Study of Hydroxyurea Use in Sickle Cell Anemia [MSH]) studies of SCD.
Materials and Methods
Patient Selection
The Walk-PHaSST Trial.
The Walk-PHaSST Trial is an Institutional Review Board–approved observational and treatment trial of cardiopulmonary disease in SCD that recruited participants with HbSS or HbSC in the United States (nine centers) and the United Kingdom (one center). Screening evaluations, all obtained as outpatients temporally separated from vaso-occlusive crises, encompassed medical history (self-reported; including medication and transfusion history), physical examination, an echocardiogram, a 6-minute walk distance (6-MWD), and routine hematologic and serologic tests. A single spot urinary albumin-to-creatinine measurement was made during the baseline visit.
These screening data served to identify participants who were eligible for the main interventional trial. The observational cohort may, therefore, have been enriched with symptomatic patients. Eligibility for treatment included both an elevated estimated pulmonary arterial systolic pressure (ePASP) on the basis of tricuspid regurgitant jet velocity (TRV) measured by echocardiogram and physiologic impairment by 6-MWD. Mortality was assessed at a median of 29 months. The analyses used data from the baseline visit and included all adult participants ≥18 years of age in the observational cohort. Informed consent was obtained for all participants in the Walk-PHaSST Trial (www.clinicaltrials.gov; NCT00492531).
Data on adults with phenotypic HbSS (≥18 years of age; including both genetic HbSS and HbS-β0 thalassemia) were examined from two additional populations.
The CSSCD Study.
CSSCD is a National Institutes of Health (NIH) –funded multicenter, observational study that included demographics, history, physical examinations, routine laboratory tests, and mortality data on 708 adults (≥18 years of age) with phenotypic homozygous HbSS between 1979 and 1988, with mortality assessments within a median of 3 years. There was no overt bias for screening. However, most participating sites were pediatric, and a greater proportion of the CSSCD population was <30 years of age compared with those in the Walk-PHaSST Trial (Figure 1). Urinary protein in this cohort was measured semiquantitatively at baseline by urinary dipstick. Only 13 adults with HbSC were enrolled in the CSSCD, and this small subgroup was excluded from these analyses.
Figure 1.
Age distribution, the Treatment of Pulmonary Hypertension and Sickle Cell Disease with Sildenafil Therapy (Walk-PHaSST) Trial, and the Cooperative Study of Sickle Cell Disease (CSSCD). Shown is the distribution of age at entry by percentage in adult subjects from the Walk-PHaSST Trial (hemoglobin SS [HbSS] and HbSC) and the CSSCD (HbSS).
The MSH Study.
MSH is an NIH–funded multicenter, placebo–controlled study of hydroxyurea use in SCD accrued through the early 1990s, and it included baseline demographics, physical examination, and laboratory data on 299 adult (≥18 years of age) participants with HbSS. This was a treatment study, and it included only patients with frequent and severe vaso–occlusive episodes.
Definitions
The estimated glomerular filtrations were calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation and the four–value Modification of Diet in Renal Disease (MDRD) equation (16–18). Microalbuminuria is a 30–300 mg/g urinary albumin-to-urinary creatinine ratio. Macroalbuminuria is a >300 mg/g urinary albumin-to-creatinine ratio. Proteinuria was reported from the CSSCD as trace or 1+–4+ from dipstick.
Statistical Analyses
Clinical and demographic characteristics were summarized using means and SDs and medians and interquartile ranges for continuous variables and percentages for categorical variables. Differences between HbSS and HbSC populations were evaluated using t test, Kruskal–Wallis rank sum test, and chi-squared test as appropriate. Multivariable linear regression was used to evaluate the characteristics associated with eGFR, albuminuria, and bicarbonate. Clinical variables incorporated into initial multivariate models, regardless of bivariate results (for the Walk-PHaSST Trial data), included age, sex, 6-MWD, TRV, complete blood count and reticulocytes, serum levels of erythropoietin (EPO), lactate dehydrogenase (LDH), potassium, creatinine, phosphorus, bicarbonate, ferritin, total bilirubin, alkaline phosphatase, albuminuria, presence of α-thalassemia mutation, BP, pulse, percentage saturation of oxygen, and use of hydroxyurea or angiotensin–converting enzyme (ACE) inhibitors. Backward stepwise regression (P<0.10) was used to evaluate the characteristics independently associated with eGFR, albuminuria, and bicarbonate, with retention of variables previously shown to be associated and selected a priori as possibly associated with each outcome and the severity of SCD. Logistic regression was used to evaluate whether eGFR and proteinuria were associated with mortality. Models were evaluated by examining residual versus fitted plots, normal Q-Q plots of standardized residuals, scale location plots, and plots of standardized residuals versus leverage.
Results
Patient Demographics and Kidney Function
In the Walk-PHaSST Trial, adults with HbSC (n=127) were older, had higher BP, had less anemia, and had lower ePASP by TRV on echocardiogram but similar physiologic reserve by 6-MWD compared with adults with HbSS (n=463) (Figure 1, Table 1).
Table 1.
Demographics in adults with sickle cell disease as mean (SD)
| Populations | SCD Phenotype, No., and Statistical Test | ||||
|---|---|---|---|---|---|
| Walk-PHaSST Trial HbSS Versus HbSC | P Value | CSSCD HbSS, n=706 | MSH HbSS, n=299 | ||
| HbSS, n=463 | HbSC, n=127 | ||||
| Age, yr | 36.6 (11.9) | 40.8 (14) | <0.001 | 27.5 (8.2) | 29.8 (7.5) |
| Sex, women | 52.1% | 59.8% | 0.12 | 56.2 | 51.2 |
| Systolic BP, mmHg | 118.7 (14) | 122.4 (15) | <0.01 | 112.5 (13.5) | 113.7 (13.5) |
| Diastolic BP, mmHg | 67.3 (10) | 74.0 (11) | <0.001 | 68.1 (10.1) | 66.0 (11.2) |
| 6-MWD, m | 439 (95) | 423 (105) | 0.11 | n/a | n/a |
| TRV, m/s | 2.6 (0.4) | 2.5 (0.4) | <0.01 | n/a | n/a |
| Hb, g/dl | 8.7 (1.6) | 11.2 (1.6) | <0.001 | 8.9 (1.4) | 8.5 (1.3) |
| LDH, IU/dl | 495 (314) | 323 (199) | <0.001 | 420 (213) | |
| Total bilirubin, mg/dl | 3.45 (3.1) | 1.96 (1.6) | <0.001 | 3.4 (2.4) | 3.7 (2.4) |
| Reticulocyte count, ×106/ml | 269 (137) | 169 (103) | <0.001 | 315 (160) | 326 (96) |
| White blood cell, ×103/µl | 10.1 (3.8) | 8.5 (3.2) | <0.001 | 11.5 (3.3) | 12.5 (3.3) |
| eGFR, ml/min per 1.73 m2 | 124.2 (39) | 111.4 (30) | <0.001 | 133.1 (34) | 120.6 (25) |
| Urine albumin-to-creatinine ratio median (IQR; n evaluable) | 71 (19–273; n=285) | 25 (8–98; n=62) | 0.001 | n/a | n/a |
| Microalbuminuria, % (n) | 43.5 (124) | 22.6 (14) | 0.002 | n/a | n/a |
| Macroalbuminuria, % (n) | 22.1 (63) | 17.7 (11) | 0.45 | n/a | n/a |
| Proteinuria measured by urine dipstick, % (n) | n/a | n/a | None: 75.8% (499); trace: 10.1% (67); 1+: 7.1% (47); 2+–4+: 6.8% (45) | n/a | |
| Bicarbonate, mEq/L | 23.8 (3.4) | 24.8 (3.3) | 0.004 | n/a | n/a |
| Potassium, mM/L | 4.46 (0.5) | 4.24 (0.5) | <0.001 | n/a | n/a |
| Phosphorus, mg/dl | 4.0 (1.6) | 3.6 (0.7) | 0.02 | 4.1 (0.6) | n/a |
| Erythropoietin, U/L | 104.2 (123) | 54.7 (51.2) | <0.001 | n/a | n/a |
| Ferritin median (IQR; n evaluable) | 294 (125–690) | 102 (59–246) | <0.001 | n/a | n/a |
| Concurrent α-thalassemia | 32.5% (137) | 31.0% (35) | 0.76 | 35.1% (248) | 32.4% (97) |
| Coincident diabetes, % (n) | 0.8 (4) | 5.5 (7) | 0.002 | 1.4 (10) | n/a |
eGFR is by the Chronic Kidney Disease Epidemiology Collaboration method. SCD, sickle cell disease; Walk-PHaSST Trial, Treatment of Pulmonary Hypertension and Sickle Cell Disease with Sildenafil Therapy Trial; HbSS, hemoglobin SS; CSSCD, Cooperative Study of Sickle Cell Disease; MSH, Multicenter Study of Hydroxyurea in Sickle Cell Disease; 6-MWD, 6-minute walk distance; n/a, not available; TRV, tricuspid regurgitant jet velocity; LDH, lactate dehydrogenase; IQR, interquartile range.
Participants with HbSS had evidence for active hemolysis (elevated LDH, total bilirubin, and reticulocyte count) and a higher EPO level than participants with HbSC. The mitigating α-thalassemia globin gene mutation was equally prevalent in patients with HbSS and patients with HbSC.
Kidney Function
Characteristics.
eGFR was high in adults with HbSS in the Walk-PHaSST Trial, the CSSCD, and the MSH (Table 1). In the Walk-PHaSST Trial, eGFR was lower among older individuals with SCD, most prominently in HbSS compared with HbSC (−1.78 ml/min per 1.73 m2 per year of age; 95% confidence interval [95% CI], −2.06 to −1.50 in HbSS and −1.09 ml/min per 1.73 m2 per year of age; 95% CI, −1.39 to 10.79 in HbSC; P<0.001). Although it may be difficult to compare this association across cohorts because of differences in comorbidities and treatments, it was of interest that eGFR was also lower among older individuals with HbSS in 706 participants in the CSSCD (−1.69 ml/min per 1.73 m2 per year of age; 95% CI, −2.00 to −1.38) and 299 participants in the MSH (−1.75 ml/min per 1.73 m2 per year of age; 95% CI, −2.05 to −1.44; n=299) (Figure 2). As a comparator in an adult nonanemic population, older age was associated with a lower measured GFR (using different techniques; −0.49 ml/min per 1.73 m2 per year of age; n=365 [19]), with a range of 0.46–0.71 ml/min per 1.73 m2 per year in men and women.
Figure 2.
Association between age and eGFR in subjects with sickle cell disease (SCD). Shown is the cross-sectional change in eGFR (milliliters per minute per year of linear regression analysis [95% confidence interval (95% CI)]) from people with hemoglobin SS (HbSS) at −1.78 (95% CI, −2.06 to −1.50), −1.72 (95% CI, −2.05 to −1.44), and −1.69 (95% CI, −2.00 to −1.38; the Treatment of Pulmonary Hypertension and Sickle Cell Disease with Sildenafil Therapy [Walk-PHaSST] Trial, the Cooperative Study of Sickle Cell Disease [CSSCD], and the Multicenter Study of Hydroxyurea in Sickle Cell Disease [MSH], respectively), with HbSC at −1.09 (95% CI, −1.39 to −0.79; the Walk-PHaSST Trial) and published non–SCD controls at −0.5 (19).
Proteinuria.
In the Walk-PHaSST Trial, two thirds of all adults with HbSS had detectable albuminuria (Table 1). Over one third of adults with HbSS from the third decade of life onward had microalbuminuria, but this prevalence was not seen in adults with HbSC until the fifth decade (Figure 3A). The proportion of adults with HbSS and macroalbuminuria was higher in older people and was greater than one third of all patients (36.7%) by the sixth decade of life (Figure 3B). Using less sensitive techniques, proteinuria was detected in one quarter of adult participants with HbSS in the CSSCD (Table 1). Despite the prevalence of renal disease in the Walk-PHaSST Trial, systemic BP was lower in adults with HbSS than in adults with HbSC (Table 1). However, this difference diminished when accounting for age and eGFR in this older, probably sicker Walk-PHaSST Trial population.
Figure 3.
Albuminuria in CKD in sickle cell disease. Shown is prevalence in percentages ≤60% by decade of age up to the sixth decade of (A) microalbuminuria and (B) macroalbuminuria in patients with hemoglobin SS (HbSS; black) or HbSC (gray).
Acid-Base.
Strikingly, bicarbonate levels were lower by approximately 1 mEq/L in adults with HbSS (Figure 4) (P=0.004) compared with those with HbSC and published reports of urban adult controls without renal disease (20); 25.4% of participants with HbSS had a strikingly low serum bicarbonate (<22 mEq/dl) compared with 14.8% of participants with HbSC (P<0.01). Serum potassium was significantly higher in HbSS compared with HbSC (P<0.001; univariate) (Table 1).
Figure 4.
Serum bicarbonate in sickle cell disease (SCD; the Treatment of Pulmonary Hypertension and Sickle Cell Disease with Sildenafil Therapy Trial). Shown is mean serum bicarbonate (ml/l±SD) in historic nonanemic controls (24.8±2.9 [1]; white bar), subjects with hemoglobin SS (HbSS; 23.8 ±3.4; gray bar; P=0.004 versus HbSC), and subjects with HbSC (24.8±3.3; black bar).
Correlates
eGFR.
In multivariate analyses of adults with HbSS in the Walk-PHaSST Trial (Table 2), age inversely associated with eGFR (P<0.001), and a low eGFR associated with cardiovascular abnormalities (a higher systolic BP [P<0.001] and a higher TRV on echocardiogram [P<0.01]) and inflammation (a higher white blood cell count; P<0.01), with evidence for tubular damage (a lower serum bicarbonate [P=0.004] and an elevated serum potassium level [P=0.01]) and a diminished erythroid reserve (a lower reticulocyte count; P=0.002). In adults with HbSC, age also inversely associated with eGFR (P<0.001), and low eGFR associated with a diminished erythroid reserve (a lower reticulocyte count; P=0.02) and use of ACE inhibitors (P<0.001). In neither subtype of SCD did a low eGFR associate with hemolysis (e.g., with a low Hb or an elevated LDH), hydroxyurea use, or concomitant α-thalassemia. Total bilirubin was directly associated with eGFR in two studies (i.e., hemolysis associating with a supranormal eGFR in the Walk-PHaSST Trial and the CSSCD) and inversely associated in a third study (the MSH). One explanation may be that patients in older studies, such as the MSH, may have been sicker and had confounding liver disease, which would increase total bilirubin, absent hemolysis, in patients with a concomitant low eGFR. Of note, results were qualitatively similar when eGFR was calculated using the CKD-EPI equation (Tables 2 and 3) or the MDRD equation (not shown).
Table 2.
Multivariate associations with eGFR (Chronic Kidney Disease Epidemiology Collaboration; hemoglobin SS and hemoglobin SC: the Treatment of Pulmonary Hypertension and Sickle Cell Disease with Sildenafil Therapy Trial)
| Variable | eGFR (ml/min per 1.73 m2) | |||
|---|---|---|---|---|
| HbSS | HbSC | |||
| Coefficient [95% CI] | P Value | Coefficient [95% CI] | P Value | |
| Age, yr | −1.78 [−2.06 to −1.50] | <0.001 | −1.09 [−1.39 to −0.79] | <0.001 |
| Men | −0.78 [−6.93 to 5.37] | 0.80 | 6.77 [−2.58 to 16.12] | 0.16 |
| Erythroid reserve/hemolytic markers | ||||
| Hb, g/dl | 1.75 [−0.31 to 3.81] | 0.10 | 1.41 [−1.77 to 4.60] | 0.39 |
| Total bilirubin,a mg/dl | 5.93 [2.3 to 9.56] | 0.002 | ||
| Absolute reticulocyte count, ×106/ml | 0.04 [0.02 to 0.07] | 0.002 | 0.06 [0.01 to 0.10] | 0.02 |
| Physiologic reserve and renal function | ||||
| TRV, m/s | −11.2 [−18.92 to −3.51] | <0.01 | ||
| Systolic BP, mmHg | −0.39 [−0.61 to −0.07] | <0.001 | ||
| Diastolic BP, mmHg | −0.12 [−0.48 to 0.23] | 0.49 | ||
| Bicarbonate, mEq/L | 1.33 [0.43 to 2.23] | 0.004 | ||
| Potassium, mM/L | −8.00 [−14.23 to −1.75] | 0.01 | −8.09 [−19.02 to 2.84] | 0.15 |
| Miscellaneous | ||||
| Alkaline phosphatase,a U/L | 2.70 [−1.70 to 7.10] | 0.23 | ||
| ACE inhibitor use | 0.62 [−8.57 to 9.80] | 0.89 | −39.0 [−57.59 to −20.34] | <0.001 |
| White blood cell count, ×106/ml | −1.22 [−2.06 to −0.38] | <0.01 | ||
| Hydroxyurea use | −2.42 [−8.64 to 3.79] | 0.44 | 3.21 [−5.53 to 11.94] | 0.47 |
| α-Thalassemia | 4.81 [−1.51 to 11.12] | 0.14 | −2.17 [−10.55 to 6.20] | 0.61 |
Blank cells indicate variables not included in the final model. HbSS, hemoglobin SS; 95% CI, 95% confidence interval; TRV, tricuspid regurgitant jet velocity; ACE, angiotensin–converting enzyme.
Per 50% higher.
Table 3.
Multivariate associations with eGFR (Chronic Kidney Disease Epidemiology Collaboration; hemoglobin SS: the Multicenter Study of Hydroxyurea in Sickle Cell Disease and the Cooperative Study of Sickle Cell Disease)
| Variable | eGFR, ml/min per 1.73 m2 | |||
|---|---|---|---|---|
| MSH | CSSCD | |||
| Coefficient [95% CI] | P Value | Coefficient [95% CI] | P Value | |
| Age, yr | −1.75 [−2.05 to −1.44] | <0.001 | −1.69 [−2.00 to −1.38] | <0.001 |
| Men | 6.40 [1.69 to 11.12] | <0.01 | 8.10 [2.74 to 13.46] | 0.001 |
| Hb, g/dl | 2.73 [0.89 to 4.57] | 0.003 | 1.29 [−0.66 to 3.25] | 0.28 |
| Total bilirubin,a mg/dl | −1.84 [−4.80 to 1.12] | 0.22 | 3.70 [0.78 to 6.63] | 0.09 |
| Absolute reticulocyte count, ×106/ml | 28.66 [3.07 to 54.25] | 0.03 | 1.63 [−13.95 to 17.21] | 0.73 |
MSH, Multicenter Study of Hydroxyurea in Sickle Cell Disease; CSSCD, Cooperative Study of Sickle Cell Disease; 95% CI, 95% confidence interval; Hb, hemoglobin.
Per 50% higher.
Albuminuria.
Unlike eGFR, albuminuria did not correlate with age in multivariate analyses when age was considered as a linear variable (Table 4). However, there was a nonlinear association between age and albuminuria in adults with HbSS only when evaluated using restricted cubic splines (P=0.004) (Figure 3B). Markers of hemolysis (increased LDH [P=0.001], increased absolute reticulocyte count [P=0.02], and weakly, a lower total Hb [P=0.07]) were associated with albuminuria. This trend was present but not significant in HbSC, except for LDH. Albuminuria associated with cardiovascular abnormalities in HbSS, such as an elevated systolic BP (P<0.001), and the use of renal protective agents (ACE inhibitors; P<0.01). Although the association between TRV and albuminuria was not significant in multivariate analyses, an elevated ePASP (a high TRV) on echocardiogram highly associated with macroalbuminuria: 51% (20 of 39) of participants with HbSS and a TRV≥3 m/s had macroalbuminuria, compared with 18% (41 of 228) of participants with a TRV<3 m/s (P<0.001).
Table 4.
Multivariate associations with albuminuria (hemoglobin SS and hemoglobin SC: the Treatment of Pulmonary Hypertension and Sickle Cell Disease with Sildenafil Therapy Trial)
| Variable | HbSS | HbSC | ||
|---|---|---|---|---|
| Coefficient [95% CI] | P Value | Coefficient [95% CI] | P Value | |
| Age, yr | 0.01 [−0.01 to 0.04] | 0.33 | 0.01 [−0.03 to 0.05] | 0.52 |
| Men | 0.23 [−0.43 to 0.88] | 0.50 | 0.32 [−0.99 to 1.62] | 0.63 |
| Erythroid reserve/hemolytic markers | ||||
| Hb, g/dl | −0.22 [−0.46 to 0.01] | 0.07 | −0.27 [−0.64 to 0.11] | 0.17 |
| LDH,a IU/L | 1.15 [0.71 to 1.59] | <0.001 | 1.37 [0.32 to 2.42] | 0.01 |
| Total bilirubin,a mg/dl | −0.61 [−1.02 to −0.20] | 0.004 | ||
| Absolute reticulocyte count, ×106/ml | <0.01 [0.00 to 0.01] | 0.02 | ||
| Physiologic reserve and renal function | ||||
| Systolic BP, mmHg | 0.05 [0.02 to 0.07] | <0.001 | ||
| Diastolic BP, mmHg | 0.07 [0.02 to 0.12] | <0.01 | ||
| Bicarbonate, mEq/L | −0.08 [−0.17 to 0.02] | 0.11 | ||
| Potassium, mM/L | 1.47 [−0.18 to 3.13] | 0.09 | ||
| Alkaline phosphatase,a U/L | 0.54 [0.10 to 0.97] | 0.02 | ||
| ACE inhibitor use | 1.25 [0.33 to 2.18] | <0.01 | −2.81 [−6.78 to 1.17] | 0.17 |
| Hydroxyurea use | −0.55 [−1.20 to 0.09] | 0.09 | 1.19 [−0.13 to 2.51] | 0.08 |
| α-Thalassemia | −0.37 [−1.03 to 0.28] | 0.27 | 0.37 [−1.00 to 1.74] | 0.60 |
Blank cells indicate variables not included in the final model. HbSS, hemoglobin SS; 95% CI, 95% confidence interval; LDH, lactate dehydrogenase; ACE, angiotensin–converting enzyme.
Per 50% higher.
Trace or greater levels of proteinuria were seen in 24.2% of the younger CSSCD population evaluated >20 years ago. This was strongly associated with a lower Hb in multivariate analyses (suggesting hemolysis; P<0.001) and more advanced age (P=0.04).
Tubular Function.
In multivariate analyses, a lower eGFR was associated with a lower serum bicarbonate (P<0.001), low total Hb (P=0.004), and an elevated EPO level (P<0.01). Despite elevated EPO levels, people with HbSS and lower serum bicarbonate did not have increased reticulocytes (instead, a trend toward a lower reticulocyte count; P=0.07), suggesting an impaired hematopoietic response to EPO. Irrespective of Hb, eGFR and EPO levels did not highly correlate in adults with HbSS in the Walk-PHaSST Trial (P=0.09). Abnormal cardiovascular reserve (Table 5) (a lower 6-MWD; P<0.001) also associated with a low bicarbonate in participants with HbSS. These changes in acid-base homeostasis were not present in participants with HbSC.
Table 5.
Multivariate associations with bicarbonate (hemoglobin SS: the Treatment of Pulmonary Hypertension and Sickle Cell Disease with Sildenafil Therapy Trial)
| Variable | HbSS | |
|---|---|---|
| Coefficient [95% CI] | P Value | |
| Age, yr | 0.03 [−0.01 to 0.06] | 0.16 |
| Men | −0.24 [−0.96 to 0.49] | 0.52 |
| Erythroid reserve/hemolytic markers | ||
| Hb, g/dl | 0.34 [0.11 to 0.57] | 0.004 |
| Erythropoietin,a U/L | −0.43 [−0.76 to −0.11] | <0.01 |
| LDH,a IU/L | ||
| Total bilirubin,a mg/dl | −0.71 [−1.12 to −0.30] | <0.001 |
| Absolute reticulocyte count, ×106/ml | <0.01 [0.00 to 0.01] | 0.07 |
| Physiologic reserve and renal function | ||
| TRV, m/s | 0.81 [−0.12 to 1.74] | 0.09 |
| 6-MWD, m | 0.01 [0.01 to 0.01] | <0.001 |
| Systolic BP, mmHg | 0.03 [0.00 to 0.05] | 0.04 |
| Potassium, mM/L | −0.87 [−1.56 to −0.18] | 0.01 |
| Miscellaneous | ||
| Hydroxyurea use | 0.96 [0.24 to 1.67] | <0.01 |
| eGFR, ml/min per 1.73 m2 | 0.01 [0.01 to 0.02] | <0.001 |
HbSS, hemoglobin SS; 95% CI, 95% confidence interval; LDH, lactate dehydrogenase; TRV, tricuspid regurgitant jet velocity; 6-MWD, 6-minute walk distance.
Per 50% higher.
Mortality
Twenty adult participants with HbSS died during 29 months median follow-up in the Walk-PHaSST Trial. Nine of 19 evaluable participants with HbSS who died had a severely elevated ePASP (TRV≥3 m/s). We cannot be conclusive about mortality. Six of seven evaluable deceased participants had albuminuria; however, definitive associations between albuminuria and mortality could not be determined because of absent measurements on 12 deceased participants.
In the CSSCD, 59 deaths occurred over approximately 3 years. After adjustment for age and eGFR, overt proteinuria was associated with higher odds of mortality in HbSS (odds ratio, 2.48; 95% CI, 1.07 to 5.77).
Discussion
Here, we described the renal phenotype seen in a sizable adult, largely North American SCD population and compared renal function between adults with HbSS and adults with HbSC. We confirmed and extended previous reports from single centers in the United States and the National Health Service in the United Kingdom (8,10,21). The strength of this report is in its derivation from the Walk-PHaSST Study, which was large, multicentered, and included a cardiovascular and renal phenotype in both participants with HbSS and participants with HbSC. We also examined historical populations to corroborate the consistency of observed associations in HbSS.
Although creatinine-based estimates of GFR in SCD may overestimate renal function (because of tubular hyperfiltration of creatinine or low muscle mass [22]), this should not significantly affect correlative relationships. The association between older age and lower eGFR seemed greatest in HbSS compared with HbSC and historic nonanemic controls. A lower eGFR in HbSS was associated with a higher TRV, with an elevated white count and a diminished erythroid reserve (reflected in a lower reticulocyte count in the Walk-PHaSST Trial and the MSH, both of which were enriched for sicker patients) regardless of whether estimated by the MDRD or the CKD-EPI method.
Unlike albuminuria (15), eGFR did not associate with hemolysis. The biochemical and metabolic effects (e.g., on bone or cardiovascular health) in SCD of a lower than normal eGFR in older individuals from a supranormal eGFR in childhood are not known but are likely to be deleterious (23).
Systemic BP has been reported to be usually (24) but not always (25) absolutely lower in adults with HbSS. Lower BP may reflect a hemodynamic response to profound anemia caused by excessive systemic vasodilation (nitric oxide and prostacyclin mediated [26,27]) and endothelial dysfunction. Impaired volume regulation caused by hyposthenuria in HbSS may also contribute (28,29). However, relative hypertension (measurements within the higher quartiles of BP for the population) or pulse pressure may be more important than absolute BP in describing vascular risk and renal morbidity in this population (30–32).
Albuminuria, in HbSS more than HbSC, was common in our population and associated with hemolysis and cardiovascular disease, such as a higher systolic BP and possibly, a higher TRV. An elevated TRV on echocardiogram highly associated with macroalbuminuria: 51% (20 of 39) of participants with HbSS and a TRV≥3 m/s had macroalbuminuria compared with 18% (41 of 228) of participants with a TRV<3 m/s (P<0.001). Although probable, we cannot confirm that albuminuria associated with a risk of mortality in the Walk-PHaSST Trial, because only one half of all deceased patients had been evaluated by urinalysis. However, six of seven participants who died during 28 months of follow-up had albuminuria. The less sensitive dipstick evaluation for proteinuria in the younger CSSCD population showed both a lower prevalence at one quarter of adult participants with HbSS and an association with mortality, irrespective of age or eGFR.
In adults with HbSC (the Walk-PHaSST Trial), we reported a higher prevalence of albuminuria than had been reported elsewhere (8). This may have reflected the severity of disease in the patients in the Walk-PHaSST Trial. In addition, earlier reports were from African-British populations only, in which mitigating α-globin gene mutations were reported in 40% of the population rather than in 30%, such as in this work.
Finally, lower serum bicarbonate levels were seen in participants with HbSS (20). This was likely multifactorial (11,33). Acidosis may be deleterious in HbSS, because it alters the oxygen dissociation curve, favoring sickling (34). In the Walk-PHaSST Trial, low serum bicarbonate levels in HbSS associated with anemia and high EPO levels, but reticulocyte counts were not increased in these patients. We speculated that erythroid precursors in an acidotic environment in SCD may be nonresponsive to EPO, which has been reported in patients without SCD on dialysis (9,35).
Our work was limited by the nature of the reported cohorts, which were not prospectively designed to examine renal phenotype. Urinalyses in the Walk-PHaSST Trial and the CSSCD were obtained only once during a baseline visit, regardless of time of day, which detracts from the rigor of the data but should not introduce systematic bias. A single clinical site in the Walk-PHaSST Trial did not collect any albumin-to-creatinine data at all; however, participants for whom albumin-to-creatinine ratios were unavailable did not differ significantly with regard to total Hb, age, LDH, reticulocyte count, or eGFR from those participants for whom we had data.
Prospective and comprehensive analyses in future in SCD may better characterize the renal phenotype. However, in the Walk-PHaSST Trial, we report on a large number of participants analyzed contemporaneously at multiple clinical sites, which reflects the current clinical status of renal disease in a modern SCD population.
We report distinct, if overlapping, complications and clinical correlates of renal disease in people with HbSS compared with those with HbSC. The distinct phenotypes likely reflect the differing degrees of hemolysis and intrarenal tubular damage and vasculopathy that are seen in HbSS compared with HbSC (36,37). At present, renal disease in SCD and its causes and complications are being inadequately and inconsistently addressed in adults, in part because of our limited recognition of its nuanced presentation. Future therapeutics may include agents aimed at interrupting renal damage that is caused by intravascular hemolysis and vaso-occlusion. Optimal management of adults should be targeted at monitoring and mitigating the acidemia (in HbSS), relative hypertension (in HbSS), the lower eGFR in older individuals, and the high prevalence of albuminuria that is found in SCD overall.
Disclosures
M.G. reports the following disclosures. Consultant: Bayer HealthCare (Whippany, NJ; reason: sickle cell disease). Payments total <$10,000 per year. This includes consulting fees and travel. Patent: Coinventor on a US Government patent for the use of nitrite salts for cardiovascular indications. Patent is owned by the government and licensed to pharmaceutical companies. Grant funding: Receives research funding from National Institutes of Health/National Heart, Lung, and Blood Institute, Aires Pharmaceuticals, Bayer HealthCare, Institute for Transfusion Medicine, and Hemophilia Center of Western Pennsylvania. This funding all goes to the University of Pittsburgh and not to M.G. directly. Authorship: Coauthor of a textbook for medical students for MedMaster, Inc. The remaining coauthors report no disclosures.
Acknowledgments
The authors acknowledge the efforts of the investigators of the Treatment of Pulmonary Hypertension and Sickle Cell Disease with Sildenafil Therapy Trial, the Cooperative Study of Sickle Cell Disease, and the Multicenter Study of Hydroxyurea in Sickle Cell Disease. We are grateful for the generous participation in these studies by people with sickle cell disease.
M.N. was supported by National Institutes of Health Research Grant P50HL118006. J.L. acknowledges institutional support from the Department of Medicine at University Hospitals Case Medical Center.
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
References
- 1.Baum KF, Dunn DT, Maude GH, Serjeant GR: The painful crisis of homozygous sickle cell disease. A study of the risk factors. Arch Intern Med 147: 1231–1234, 1987 [PubMed] [Google Scholar]
- 2.Ohene-Frempong K, Weiner SJ, Sleeper LA, Miller ST, Embury S, Moohr JW, Wethers DL, Pegelow CH, Gill FM: Cerebrovascular accidents in sickle cell disease: Rates and risk factors. Blood 91: 288–294, 1998 [PubMed] [Google Scholar]
- 3.Platt OS, Thorington BD, Brambilla DJ, Milner PF, Rosse WF, Vichinsky E, Kinney TR: Pain in sickle cell disease. Rates and risk factors. N Engl J Med 325: 11–16, 1991 [DOI] [PubMed] [Google Scholar]
- 4.Powars DR, Chan LS, Hiti A, Ramicone E, Johnson C: Outcome of sickle cell anemia: A 4-decade observational study of 1056 patients. Medicine (Baltimore) 84: 363–376, 2005 [DOI] [PubMed] [Google Scholar]
- 5.Powars DR, Hiti A, Ramicone E, Johnson C, Chan L: Outcome in hemoglobin SC disease: A four-decade observational study of clinical, hematologic, and genetic factors. Am J Hematol 70: 206–215, 2002 [DOI] [PubMed] [Google Scholar]
- 6.Nagel RL, Fabry ME, Steinberg MH: The paradox of hemoglobin SC disease. Blood Rev 17: 167–178, 2003 [DOI] [PubMed] [Google Scholar]
- 7.Thomas AN, Pattison C, Serjeant GR: Causes of death in sickle-cell disease in Jamaica. Br Med J (Clin Res Ed) 285: 633–635, 1982 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Sharpe CC, Thein SL: How I treat renal complications in sickle cell disease. Blood 123: 3720–3726, 2014 [DOI] [PubMed] [Google Scholar]
- 9.Diskin CJ, Stokes TJ, Dansby LM, Radcliff L, Carter TB: Can acidosis and hyperphosphataemia result in increased erythropoietin dosing in haemodialysis patients? Nephrology (Carlton) 11: 394–399, 2006 [DOI] [PubMed] [Google Scholar]
- 10.Ataga KI, Orringer EP: Renal abnormalities in sickle cell disease. Am J Hematol 63: 205–211, 2000 [DOI] [PubMed] [Google Scholar]
- 11.Batlle D, Itsarayoungyuen K, Arruda JA, Kurtzman NA: Hyperkalemic hyperchloremic metabolic acidosis in sickle cell hemoglobinopathies. Am J Med 72: 188–192, 1982 [DOI] [PubMed] [Google Scholar]
- 12.Guasch A, Cua M, Mitch WE: Early detection and the course of glomerular injury in patients with sickle cell anemia. Kidney Int 49: 786–791, 1996 [DOI] [PubMed] [Google Scholar]
- 13.Steinberg MH, Barton F, Castro O, Pegelow CH, Ballas SK, Kutlar A, Orringer E, Bellevue R, Olivieri N, Eckman J, Varma M, Ramirez G, Adler B, Smith W, Carlos T, Ataga K, DeCastro L, Bigelow C, Saunthararajah Y, Telfer M, Vichinsky E, Claster S, Shurin S, Bridges K, Waclawiw M, Bonds D, Terrin M: Effect of hydroxyurea on mortality and morbidity in adult sickle cell anemia: Risks and benefits up to 9 years of treatment. JAMA 289: 1645–1651, 2003 [DOI] [PubMed] [Google Scholar]
- 14.Maurel S, Stankovic Stojanovic K, Avellino V, Girshovich A, Letavernier E, Grateau G, Baud L, Girot R, Lionnet F, Haymann JP: Prevalence and correlates of metabolic acidosis among patients with homozygous sickle cell disease. Clin J Am Soc Nephrol 9: 648–653, 2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Saraf SL, Zhang X, Kanias T, Lash JP, Molokie RE, Oza B, Lai C, Rowe JH, Gowhari M, Hassan J, Desimone J, Machado RF, Gladwin MT, Little JA, Gordeuk VR: Haemoglobinuria is associated with chronic kidney disease and its progression in patients with sickle cell anaemia. Br J Haematol 164: 729–739, 2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D, Modification of Diet in Renal Disease Study Group : A more accurate method to estimate glomerular filtration rate from serum creatinine: A new prediction equation. Ann Intern Med 130: 461–470, 1999 [DOI] [PubMed] [Google Scholar]
- 17.Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF, 3rd, Feldman HI, Kusek JW, Eggers P, Van Lente F, Greene T, Coresh J, CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) : A new equation to estimate glomerular filtration rate. Ann Intern Med 150: 604–612, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Arlet JB, Ribeil JA, Chatellier G, Eladari D, De Seigneux S, Souberbielle JC, Friedlander G, de Montalembert M, Pouchot J, Prié D, Courbebaisse M: Determination of the best method to estimate glomerular filtration rate from serum creatinine in adult patients with sickle cell disease: A prospective observational cohort study. BMC Nephrol 13: 83, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rule AD, Gussak HM, Pond GR, Bergstralh EJ, Stegall MD, Cosio FG, Larson TS: Measured and estimated GFR in healthy potential kidney donors. Am J Kidney Dis 43: 112–119, 2004 [DOI] [PubMed] [Google Scholar]
- 20.Shah SN, Abramowitz M, Hostetter TH, Melamed ML: Serum bicarbonate levels and the progression of kidney disease: A cohort study. Am J Kidney Dis 54: 270–277, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sharpe CC, Thein SL: Sickle cell nephropathy - a practical approach. Br J Haematol 155: 287–297, 2011 [DOI] [PubMed] [Google Scholar]
- 22.Asnani MR, Reid ME: Renal function in adult Jamaicans with homozygous sickle cell disease. Hematology 20: 422–428, 2015 [DOI] [PubMed] [Google Scholar]
- 23.Miller JA, Curtis JR, Sochett EB: Relationship between diurnal blood pressure, renal hemodynamic function, and the renin-angiotensin system in type 1 diabetes. Diabetes 52: 1806–1811, 2003 [DOI] [PubMed] [Google Scholar]
- 24.Thompson J, Reid M, Hambleton I, Serjeant GR: Albuminuria and renal function in homozygous sickle cell disease: Observations from a cohort study. Arch Intern Med 167: 701–708, 2007 [DOI] [PubMed] [Google Scholar]
- 25.Desai PC, Deal AM, Brittain JE, Jones S, Hinderliter A, Ataga KI: Decades after the cooperative study: A re-examination of systemic blood pressure in sickle cell disease. Am J Hematol 87: E65–E68, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Kaul DK, Zhang X, Dasgupta T, Fabry ME: Arginine therapy of transgenic-knockout sickle mice improves microvascular function by reducing non-nitric oxide vasodilators, hemolysis, and oxidative stress. Am J Physiol Heart Circ Physiol 295: H39–H47, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lamarre Y, Hardy-Dessources MD, Romana M, Lalanne-Mistrih ML, Waltz X, Petras M, Doumdo L, Blanchet-Deverly A, Martino J, Tressières B, Maillard F, Tarer V, Etienne-Julan M, Connes P: Relationships between systemic vascular resistance, blood rheology and nitric oxide in children with sickle cell anemia or sickle cell-hemoglobin C disease. Clin Hemorheol Microcirc 58: 307–316, 2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Itano HA, Keitel HG, Thompson D: Hyposthenuria in sickle cell anemia: A reversible renal defect. J Clin Invest 35: 998–1007, 1956 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Cochran RT, Jr.: Hyposthenuria in sickle cell states. Arch Intern Med 112: 222–225, 1963 [DOI] [PubMed] [Google Scholar]
- 30.Novelli EM, Hildesheim M, Rosano C, Vanderpool R, Simon M, Kato GJ, Gladwin MT: Elevated pulse pressure is associated with hemolysis, proteinuria and chronic kidney disease in sickle cell disease. PLoS One 9: e114309, 2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Gordeuk VR, Sachdev V, Taylor JG, Gladwin MT, Kato G, Castro OL: Relative systemic hypertension in patients with sickle cell disease is associated with risk of pulmonary hypertension and renal insufficiency. Am J Hematol 83: 15–18, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Pegelow CH, Colangelo L, Steinberg M, Wright EC, Smith J, Phillips G, Vichinsky E: Natural history of blood pressure in sickle cell disease: Risks for stroke and death associated with relative hypertension in sickle cell anemia. Am J Med 102: 171–177, 1997 [DOI] [PubMed] [Google Scholar]
- 33.Scheinman JI: Sickle cell disease and the kidney. Semin Nephrol 23: 66–76, 2003 [DOI] [PubMed] [Google Scholar]
- 34.Ueda Y, Nagel RL, Bookchin RM: An increased Bohr effect in sickle cell anemia. Blood 53: 472–480, 1979 [PubMed] [Google Scholar]
- 35.Ueda Y, Bookchin RM: Effects of carbon dioxide and pH variations in vitro on blood respiratory functions, red blood cell volume, transmembrane pH gradients, and sickling in sickle cell anemia. J Lab Clin Med 104: 146–159, 1984 [PubMed] [Google Scholar]
- 36.Statius van Eps LW, Schouten H, Haar Romeny-Wachter CC, La Porte-Wijsman LW: The relation between age and renal concentrating capacity in sickle cell disease and hemoglobin C disease. Clin Chim Acta 27: 501–511, 1970 [DOI] [PubMed] [Google Scholar]
- 37.Gladwin MT, Vichinsky E: Pulmonary complications of sickle cell disease. N Engl J Med 359: 2254–2265, 2008 [DOI] [PubMed] [Google Scholar]