Pediatric Nephrology
5th Edition

48
Sickle Cell Nephropathy
Jon I. Scheinman
The major clinical consequences of sickle cell disease (SCD) are vascular obstruction by sickled cells and anemia because of red blood cell (RBC) destruction. The sickling process may cause hematuria, renal papillary necrosis (RPN), and a urinary concentrating defect. There also is a chronic sickle cell glomerulopathy, which is less directly related to sickling, as well as unusual susceptibility to infections and to a recently reported specific form of malignancy.
Sickle hemoglobin (HbS) differs from normal hemoglobin (HbA) by the substitution of valine for glutamine in the 6 position of the β-globin chain (1). Under low oxygen tension in concentrated solution, HbS aggregates and forms intracellular fibers. These fibers prevent RBCs from deforming normally, so they do not pass easily through the microcirculation and thus cause occlusion. Repeated cycles make some cells irreversibly sickled, and even in well-oxygenated blood they cause abnormal viscosity. In addition, these abnormal cells are easily lysed, which causes anemia.
HEMATURIA AND RENAL PAPILLARY NECROSIS
Clinical Features
Gross hematuria may be the most dramatic clinical event in SCD (2). It is often painless and usually unilateral, more commonly on the left side (3), which is explained by increased venous pressure due to the greater length of the left renal vein. Hematuria can occur at any age. It is more often reported with sickle trait (HbAS), probably because of the higher genetic frequency of HbAS than of HbSS (4). RPN is usually discovered in patients with painless gross hematuria (5). Hematuria is not invariably present, however, and RPN can occur even in young children (3). When it was sought systematically, 40% of patients in one Nigerian series were found to have RPN (6). In another series, there was no difference in the incidence of RPN in symptomatic patients (65%) and in asymptomatic (62%) patients (7). The frequency of RPN on urography therefore suggests that the process develops subclinically without gross hematuria.
Pathogenesis
The pathology associated with isolated hematuria is rarely examined. In the past, kidneys removed from patients with uncontrolled bleeding in SCD showed relatively insignificant changes, primarily medullary congestion (4). The hematuria probably results from the sequence of renal medullary sickling, vascular obstruction, and RBC extravasation. This is precipitated by those factors present in the renal medulla that lead to sickling: The PaO2 (35 to 40 mm Hg) is below the threshold (45 mm Hg) for sickling (8); the high osmolality of the medulla draws water from the RBC, which leaves the HbS concentrated, promoting the formation of hemoglobin polymers. The acidic environment of the renal medulla further increases the likelihood of sickling. This mechanism of microvascular occlusion explains many of the complications of SCD but is overly simplistic. The distortion of the cell is also partly explained by dehydration of the cell by enhanced KCl co-transport, induced by cell swelling and acidification (9). Furthermore, K+ and water efflux are enhanced by transiently increased SS cell cytosolic Ca, induced by the membrane distortion.
Pathology
The pathology of RPN in SCD is a focal process, with some collecting ducts surviving within a diffuse area of fibrosis. Within the medullary fibrosis the vasa recta are destroyed, after initial dilation and engorgement (10). The dependence of the papilla on that circulation results from repeated small focal infarctions of the papilla (11). This differs from the RPN found in analgesia abuse, in which the vasa recta typically are spared, and most lesions occur in peritubular capillaries (11). These same factors are present in sickle trait, although with a smaller proportion of sickling cells. Because calyces are affected separately and sequentially in SCD, acute obstruction and renal failure are uncommon (7).
Diagnosis
Continued gross hematuria likely represents a form of renal “sickle crisis” in a patient with known HbSS or HbAS. Other
P.918

treatable causes of hematuria, including the distinctive renal medullary carcinoma in patients with sickle hemoglobin, must be excluded (12). Severe pain makes the diagnosis of renal sickle crisis less likely, whereas moderate discomfort often lateralizes the bleeding. Renal and bladder ultrasonography can rule out bleeding from a stone or tumor and diagnose RPN (see later). The increased echodensity of medullary pyramids on ultrasonography is typical of SCD, and in the absence of hypercalciuria, medullary echodensity in a patient with hematuria should suggest a sickle hemoglobinopathy (13). Walker (14) reviewed ultrasonographic reflectivity in young SCD patients (age 10 to 20 years) and found diffuse echogenicity in 9% and medullary echodensity in 3%. Surprisingly, these findings were greater in the milder genotypes, 37% in SC patients and 79% in Sβ+ thalassemia patients. These findings are unexplained but are unlikely to represent RPN. The echodensity was found overall in 20% of patients with sickling processes and was interpreted to suggest subclinical nephrocalcinosis or iron deposition.
FIGURE 48.1. Tomographic pyelography of an 18-year-old patient with abdominal pain and hematuria. Papillary necrosis is evident from blunted medullary cavities, especially the upper pole. The bases of the calyces are preserved. The middle pole calyx has a possible sinus tract.
The diagnosis of RPN in SCD was traditionally made by urography. In the series of McCall et al., 39% of 189 patients had calyceal clubbing, including 23% with definite RPN (15). Cortical scarring as found in pyelonephritis does not accompany the calyceal clubbing of SCD (16). Other urographic findings of RPN in SCD are distinctive. A “medullary” form in which an irregular medullary cavity is present, often with sinus tracts, is common (7). The base of the calyx and its normal outline are preserved (16). Sonography can sometimes identify the early medullary form of papillary necrosis (17). A later finding is calcification of the medullary pyramids in a “garland” pattern surrounding the pelvis. This pattern of “shadowing” echodensity may be distinctive. The progression to the “papillary” form results in clubbing and caliectasis (Figs. 48.1 and 48.2). This is more common in analgesic nephropathy, in which an area of sequestration is often found, which results from infarction of a large area of the papilla (18). A prospective survey of symptomatic SCD patients (7) found that 11 of 18 SA patients had a form of RPN, but 8 of these 11 had evidence of infection. Nine of 11 symptomatic SS patients had RPN, of which 5 had evidence of infection. Asymptomatic patients included 16 of 22 SS patients with RPN, 1 of 3 SA patients, 3 of 4 SC patients, and 5 of 8 Sβ+ patients. It is probably not necessary to perform contrast urography to visualize the renal architecture in SCD.
FIGURE 48.2. Ultrasonographic visualization of the same kidney as in Figure 48.1. The middle pole exhibits deep extensions into the papilla, likely sinus tracts, typical of the “papillary” form of renal papillary necrosis.
Treatment
In view of the benign pathology in SCD hematuria, conservative management is appropriate (4). Bed rest is often recommended to avoid dislodging hemostatic clots. It is advisable to maintain high rates of urine flow by both hypotonic fluid intake (4 L/1.73 m2 surface area per day) and administration
P.919

of diuretics (a thiazide or a loop diuretic such as furosemide), which should help clear clots from the bladder (3). In addition, a diuresis reduces medullary osmolarity and may therefore help alleviate sickling in the vasa recta. Some caution is necessary in the administration of sodium-containing fluids because of potential sodium retention (see later).
The combination of vasopressin and administration of hypotonic fluid to reduce plasma osmolality has been suggested. Hyponatremia induces water uptake by RBCs in vitro, thereby reducing the effective HbS concentration and making the cells less likely to sickle (4). This treatment is neither safe nor proven effective (4). Saline volume expansion would be especially ineffective and, coupled with hypertransfusion, would predispose to congestive heart failure. It would also be surprising if vasopressin were to increase the patient’s urine concentration sufficiently to inhibit water diuresis (see later).
Because sickling is increased in an acid environment, alkalinization of the urine by 8 to 12 g NaHCO3 (per 1.73 m2) per day may reduce sickling in a urine environment, but this may not be relevant for medullary sickling (3). Alkalinizing the patient to increase the oxygen affinity of hemoglobin is theoretically valid but not of proven value in practice (4). Transfusions may be necessary for blood loss and could be helpful by increasing the proportion of normal HbAA cells and thus reducing sickling cells.
Epsilon aminocaproic acid (EACA) inhibits fibrinolysis, allowing clots to mediate hemostasis. The effective dosage in an adult is 8 g/day. In one series, 4 of 12 cases required EACA after failure of treatment with fluid administration and alkalinization (3). Unless other measures have failed, the risk of thrombosis should impose some caution in use of EACA. Lower dosages may be adequate to arrest hematuria, starting with 1 g (per 1.73 m2) orally three times daily and increasing the dose until hemostasis occurs (4). Nephrectomy is rarely required for uncontrolled bleeding. Arteriographic localization and local embolization of the involved renal segment may avoid nephrectomy.
RPN can be prevented experimentally by either diabetes insipidus or water diuresis, which thereby eliminates the medullary concentration gradient (19). Thus the fluid administration prescribed for gross hematuria is also appropriate to prevent RPN. Angiotensin-converting enzyme (ACE) inhibition can experimentally induce a 50% increase in papillary blood flow (19). This may help prevent RPN, but it could aggravate hematuria acutely by increasing blood flow to a bleeding area.
TUBULAR DYSFUNCTION
The relevant medullary pathology in SCD is found in the region of the collecting ducts, the inner medulla, and the papilla. This results in dysfunction of the collecting duct and juxtamedullary nephrons (19). Sickling and vascular congestion of the medulla, seen in gross hematuria (4), are probably responsible for reversible concentrating defects. The basis of irreversible medullary dysfunction in RPN is likely the medullary fibrosis and destruction of vasa recta. Juxtamedullary nephrons and collecting ducts are destroyed, as is found in the model of RPN induced by bromoethylene-hydrobromide in rats (20).
Clinicopathologic Features
Urinary Concentration
The most common tubular abnormality in SCD is a urinary concentrating defect. Typically, HbSS patients achieve a urine concentration of 414 mOsm/kg after 8 to 10 hours of thirst, compared with 911 mOsm/kg in controls (21). Patients with sickle trait may have a diminished urinary concentrating capacity (22). This concentrating defect in children with SCD may result in enuresis (1) and an increased risk of dehydration during water deprivation.
The ability to concentrate urine depends on an intact collecting duct. The collecting ducts of juxtamedullary nephrons extend deepest into the medulla and are capable of generating the highest urine concentration. Poor sodium reabsorption in the collecting duct can result if the sluggish blood flow does not remove the reabsorbed sodium (19). The continued low-grade sickling and medullary congestion result in loss of the normal medullary concentration necessary for water reabsorption. This defective urine concentration can be transiently reversed by transfusion (23,24). The permanent destruction of collecting ducts by medullary fibrosis in humans with SCD (10) and in rats (20) results in an irreversible concentrating defect.
Vasopressin generation is normal in SCD, and the concentrating defect is not responsive to vasopressin. The concentrating defect is unique to sickling hemoglobinopathies, in that there is no concentrating defect in other anemias (25).
Diluting Capacity
Urinary dilution depends on the solute reabsorption in the ascending loop of Henle of cortical nephrons, which are not involved in SCD patients. They can usually dilute the urine normally (21,23).
Hydrogen Ion and Potassium Excretion
A proton gradient from tubular cell to lumen underlies acid excretion. Proton secretion is associated with the “intercalated” collecting duct cells, most prominent in the cortical segment of the collecting duct (19). Damage to the papillary segment is therefore unlikely to cause a severe acidification defect. However, juxtamedullary nephrons, which reabsorb HCO3-, are also severely involved in SCD. On this basis, some defect in acid excretion may occur (19).
An incomplete distal renal tubular acidosis (RTA) may complicate SCD, but it is usually not a clinical problem (23). The
P.920

minimum urine pH achieved in response to NH4Cl loading is not as low as in controls (5.8 vs. 5.1), but total NH4 excretion is normal. Consequently, titratable acidity is reduced (26). Kurtzman has also described a “type IV RTA” in SCD, with a reduced ability to lower urine pH in response to Na2SO4 and inadequate K+ secretion, especially in patients with decreased renal function (27). In one series, six of nine nephrotic SCD patients were reported to have type IV RTA (28).
Plasma renin and aldosterone levels may be increased in the face of medullary fibrosis (23). A protective mechanism likely exists: in the presence of inadequate K+ secretion, a shift of K+ to intracellular compartments probably occurs. Because this shift is under β2 stimulation, beta-blockers or ACE inhibition may result in hyperkalemia (23). The electrolyte abnormalities resemble those in type IV RTA but actually result from an aldosterone-independent end-organ failure secondary to medullary fibrosis.
Proximal Tubular Reabsorption
An increased capacity for sodium reabsorption and a decreased sodium excretion with loop diuretics is seen in SCD (21,29). De Jong and Statius van Eps have proposed that the alterations in renal cortical function are adaptive, compensating for defects in medullary sodium and water conservation (10). The increased proximal sodium reabsorption results in decreased distal sodium delivery. Diuretic response is poor, because it depends on this more distal sodium delivery. Proximal tubular phosphate reabsorption, which usually parallels sodium reabsorption, is also increased. This may cause hyperphosphatemia, especially in the presence of an increased phosphate load generated by hemolysis (23).
Tubular Secretion
A significant disparity is found between creatinine clearance (CCr) and inulin clearance (CIn) in SCD. For example, in a small group of patients with SCD and an increased glomerular filtration rate (GFR) as measured by CIn (119 mL/min versus 97 mL/min for individuals without SCD), CCr was significantly higher than CIn in SCD patients (154 versus 119 mL/min) but not in those without SCD (114 vs. 97 mL/min) (21). This is an expression of increased tubular secretion of creatinine in SCD (23).
Uric acid secretion is similarly increased and is a functional adaptation to high uric acid generation (5). In patients with decreasing total GFR (and increasing GFR per nephron), fractional excretion of urate is further increased by decreased reabsorption (30).
Experimental Models of Tubular Dysfunction in Sickle Cell Disease
The rat model of RPN induced by bromoethylene-hydrobromide exemplifies distal tubule physiologic disturbances (20). As a baseline, these rats have a more dilute urine than controls. The juxtamedullary nephrons are nonfunctional. Sodium excretion changes little in response to hypervolemia. Although the measured atrial natriuretic peptide (ANP) level increases appropriately, an additional high-dose infusion of ANP can generate a normal response. The normal response to hypervolemia is likely mediated through inner medullary interstitial pressure, damaged in this model (and in SCD). In contrast, the response to salt loading generates a normal increased sodium excretion in experimental RPN, regulated by a more cortical mechanism (20).
Role of Prostaglandins in Tubular Dysfunction in Sickle Cell Disease
A series of studies begun by de Jong and Statius van Eps explored the effects of prostaglandin (PG) inhibition on renal function (10,31). The effect of PG inhibition on tubular function is especially revealing: there was a greater fall (42%) in the fractional excretion of Na in response to PG inhibition by indomethacin in SCD patients than in individuals without SCD (16%). This reflects a greater than normal effect of PGs on the delivery of sodium to the distal diluting segment. Although normal urinary dilution is not affected by PG inhibition, PG inhibition decreases urinary dilution in SCD (21).
PGs increase proximal sodium reabsorption. Under PG inhibition, more solute is delivered to and then reabsorbed by the thick ascending limb of the loop of Henle, which thereby increases interstitial hypertonicity. More free water is then absorbed in the relatively solute-impermeable descending limb, which results in a decreased response to water loading (21). Thus, although urinary diluting capacity is normal in SCD, it is being maintained only by PG and will be decreased by indomethacin (10). PG inhibition increases distal delivery, preventing the appropriate effect of vasopressin suppression during water diuresis in SCD (10). Increased proximal sodium reabsorption in SCD results in a decreased natriuretic response to loop diuretics (20), and PG inhibition restores that response.
Unlike in those without SCD, in SCD patients net acid excretion fails to increase in response to inhibition of PG synthesis by indomethacin (26) because of decreased NH4+ excretion. It is likely that NH4+ excretion is maintained at a maximum by endogenous PGs.
In summary, the tubular dysfunction of SCD manifests a defect in urine concentration, while dilution is maintained. Hydrogen ion and potassium secretion functions are only mildly affected, and proximal tubular mechanisms are exaggerated.
Treatment
Treatment of tubular disorders in SCD is usually unnecessary if renal function is normal. The risk of dehydration
P.921

caused by decreased urinary concentrating ability requires earlier treatment of diarrhea or vomiting. There should be a cautious approach to volume expansion as treatment for sickle crises; administration of large volumes of standard sodium-containing fluids to significantly anemic patients with increased sodium reabsorption may result in congestive heart failure. Acidosis may require earlier treatment in the patient with SCD. Hyperuricemia, resulting from increased urate production, may be aggravated by diuretics (especially thiazides) that inhibit urate secretion. The edema accompanying severe anemia may be difficult to treat because the response to diuretics is diminished. Severe hemolysis may exceed the patient’s ability to excrete potassium, especially if there is renal insufficiency. Beta-blockers or ACE inhibition can aggravate hyperkalemia (23), especially in the presence of some degree of renal impairment (27).
SICKLE CELL GLOMERULOPATHY
Clinical Features
The association of significant proteinuria with SCD has been recognized sporadically and usually described as a nephritic process. As early as 1959 (32), nephrotic syndrome was recognized in SCD. Proteinuria was identified in 17 of 54 patients (32). A population study found proteinuria in 20% of 284 patients at a single center, with a prevalence of 29% in adults but only 5% in children younger than 10 years of age (33). Bakir et al. recognized nephrotic syndrome in 12 of 240 adults with SCD, with proteinuria of 2 to 20 g/24 hr (28). In our series, 87 (26%) of 381 adult patients had significant proteinuria, with 12 in the nephrotic range (more than 2.5 g/24 hr) (34). In a more recent study of 34 adult patients with SCD, 7 had albuminuria with normal GFR and 17 had chronic renal failure, with glomerular injury and loss of ultrafiltration coefficient (35). In this series, GFR was related to hematocrit (36).
Lonsdorfer et al. (37) found that 40 to 45% of 31 SS patients aged 16 to 40 years had abnormal proteinuria, mostly selective (albumin). Twelve percent of the 17 patients who were younger than age 16 had abnormal proteinuria. Of 52 SA patients older than age 16, approximately 18% had abnormal proteinuria. Those with Sβ+ were similar to SS patients, whereas those with SC disease fell between SS and SA patients.
The definition of sickle cell nephropathy that is most accepted is associated with nephrotic-range proteinuria. Although long-term studies have not been done, it appears to have a more rapid course than nephrotic syndrome due to other causes. In the experience of Bakir et al. (28), two-thirds of patients developed renal failure within 2 years. The onset of renal failure was heralded by increasingly inadequate erythropoiesis (38), and survival time after diagnosis was 4 years.
Pathology
The usual finding in sickle cell nephropathy is focal segmental glomerulosclerosis (FSGS), which is intimately associated with glomerular hypertrophy. Glomerular engorgement and hypertrophy were recognized as part of SCD by Berman and Tublin in 1960 (32). FSGS was reported in a 9-year-old nephrotic SCD patient in 1959 (32). Ten of our adult HbSS patients without significant renal impairment underwent renal biopsy because of proteinuria (34). Eight had FSGS involving a mean of 27% of glomeruli, and the other two had focal global sclerosis. There was focal tubulointerstitial fibrosis adjacent to sclerotic glomeruli. The nonsclerotic glomeruli were all enlarged, with diameters of 186 ± 14.5 μm versus 137.9 ± 19.3 μm in ten control biopsies. Immunofluorescence gave positive results only for immunoglobulin M, C3, and C1q irregularly in sclerotic segments. Electron microscopy confirmed the absence of immune complex–type dense deposits. There was focal electron-lucent expansion of the subendothelial zone in six specimens, with occasional mesangial cell interposition. No new mesangial matrix material was observed to suggest membranoproliferative glomerulonephritis (MPGN).
These findings agree with the description by Zamurovic and Churg (39). Glomerular hypertrophy was documented previously in adults with measured glomerular diameter (median, 257 μm; range, 220 to 316 μm) greater than in controls (median, 193 μm; range, 142 to 253 μm) (40). Bakir et al. described both the nonimmune MPGN-like lesion in nine patients, and FSGS (in 8% of glomeruli) or global sclerosis (in 14% of glomeruli) (28). In addition to FSGS, focal cortical infarcts have been described as a late finding in sickle cell nephropathy (8).
A report on the renal pathology in six HbSS patients with proteinuria described two patterns of FSGS: a “collapsing” and an “expansive” sclerosis (41). Glomeruli were significantly hypertrophied, with diameters of 233 ± 25.3 μm, compared with 158 ± 12.7 μm in subjects without SCD. Glomeruli were more hypertrophied in HbSS disease than in idiopathic FSGS (188.2 ± 17.9 μm). The glomeruli from HbSS patients without clinical evidence of renal disease (243.5 ± 12.5 μm) were the same size as in those with proteinuria.
Renal Function in Sickle Cell Glomerulopathy
Hyperfiltration
An increased GFR has been a recognized feature of SCD, especially in children, in studies dating from the 1950s (10). Renal plasma flow, as estimated by p-aminohippurate (PAH) clearance, is elevated in excess of GFR, which results in a lower than normal filtration fraction (12.9 vs. 17.5) (21). The extraction ratio of PAH (normally 90 to 95%) is also lower (24). The cause and mechanism of this physiologic
P.922

alteration are unclear, although it is likely that increased cortical blood flow itself can cause decreased secretion by limiting diffusion from rapidly flowing plasma (24). A more recent analysis suggests a distinctive pattern of increased glomerular permeability (to dextrans) in SCD nephropathy, an increase in pore radius, which is not explained by purely hemodynamic changes. When chronic renal failure develops, the total number of membrane pores is reduced and a size-selectivity defect occurs.
Role of Prostaglandins in Glomerulopathy
The suggestion has been made that hyperfiltration and proximal tubular “hyperfunction” in SCD are a compensation for the distal tubular injury, mediated by the PG systems (10,20). In the studies by de Jong and Statius van Eps (see Tubular Dysfunction) (10), indomethacin decreased GFR and the estimated renal plasma flow (ERPF) in SCD patients but did not alter that of controls, which suggests that the PG system might be responsible for maintaining the GFR in SCD. Measured prostaglandin E1 (PGE1) excretion did not differ from that in controls, but prostaglandin F2 (PGF2) excretion was lower (42), which thus increased the ratio of the vasodilator PGE1 to the vasoconstrictor PGF2. Allon et al. (21) found that indomethacin decreased PGF1 (prostacyclin, a vasodilator hormone) more in SCD patients (46%) than in control subjects (15%). GFR was decreased 16% by indomethacin in SCD but was unchanged in control subjects. The more dramatically increased ERPF and decreased filtration fraction of SCD were returned toward normal, and the net effect of PG inhibition was to reverse hyperfiltration.
Pathogenesis
Immunopathogenetic Mechanisms
In 1975, Strauss et al. (43) described an SCD membranoproliferative (MPGN) nephropathy in seven patients—four with nephrotic syndrome, and three of these under 15 years of age—with immunopathologic studies suggesting an immunocomplex disease. One proposed cause for the sickle cell nephropathy has been immunologic reaction to renal tubular epithelial cell complexes (10). Most now agree that evidence of immunocomplex deposition is usually lacking in SCD patients with heavy proteinuria.
Relationship to Other Glomerulopathies
The relationship of the pathologic findings of SCD to those of other nephropathies is unclear. Glomerular hypertrophy with FSGS is also found in the setting of reduced renal mass. Examples are reflux nephropathy, severe obesity (44), and rat models of renoprival glomerulosclerosis (45). In FSGS associated with idiopathic nephrotic syndrome, the glomeruli that develop sclerosis are hypertrophied (1). The nephropathy associated with type 1 glycogen-storage disease (GSD-1) includes both hyperfiltration and FSGS (46). The nephropathy associated with cyanotic congenital heart disease may arise from mechanisms similar to those operating in SCD: low oxygenation and increased blood viscosity from polycythemia (23,47).
Possible Mechanisms of Sickle Cell Disease Nephropathy
Hyperfiltration, in concert with direct endothelial damage by occlusion with sickled cells, might lead to endothelial hyperplasia and ultimately fibrosis (10,23). The iron deposited in tubular cells as hemosiderin has been suspected to have a role in the chronic nephropathy of SCD (8). The non–immune complex deposits found in SCD might derive from iron–protein complexes (8). Experimentally, saturated-iron complexes can induce a nephrotic syndrome in rabbits (24). Lande et al. found decreased renal cortical spin-echo signal by magnetic resonance imaging in SCD, which suggests an abnormal renal cortical iron metabolism (48). This does not occur in β-thalassemia, despite similar iron overload.
It is possible that FSGS is the consequence rather than the cause of interstitial fibrosis, which might obstruct the efferent glomerular capillaries, raising intraglomerular pressure and resulting in progressive (reactive) sclerosis (49). In SCD, in which medullary fibrosis is most prominent, the vasa recta supplying the juxtamedullary nephrons would be most affected by FSGS. Bhathena et al. (41) suggested that the “collapsing” pattern of FSGS, superimposed on already maximally hypertrophied glomeruli, might be a consequence of sickling and ischemic collapse, similar to the glomerular “microinfarcts” suggested by Chauhan et al. (8). The “expansive” form of FSGS is viewed as a mesangial cell reaction to capillary collapse.
Both the pathophysiologic and pathologic findings in sickle cell nephropathy resemble those in the rodent model of glomerular hypertension induced by renal mass reduction (50). Thus hyperfiltration, glomerular enlargement, and focal and segmental glomerulosclerosis could be a result of an increase in intraglomerular pressure as a consequence of efferent arteriolar vasoconstriction. In that model, and in others, the glomerular hypertension and the pathologic consequences are attenuated by angiotensin II–converting enzyme inhibition.
Any hypothesis should take into account the glomerular hypertrophy that is always present in SCD, probably related to the anemia itself. Proteinuria is not invariable in SCD and appears to be unrelated to the number and severity of SCD crises or the presence of hematuria, RPN, and so on. Therefore some other factors are likely to be operative in those at greatest risk of proteinuria and FSGS. Systemic hypertension is notably absent in these patients (34). The presence of hyperfiltration, glomerular hypertrophy,
P.923

and FSGS does not imply that these findings are sequential or causative. A common stimulus may be operative, such as the growth-promoting hormones and cytokines (57), to which the glomerulus may be sensitive.
Diagnosis
Hyperfiltration
The upper limit of the normal range for GFR is not certain, even with CIn, the gold standard. The reliability of the clearance methods that can substitute for CIn have not been validated in the elevated range. In adult SCD subjects aged 40 to 75 years, CCr correlated well with the clearance of chromium-51 ethylenediaminetetra-acetic acid (51Cr-EDTA) when clearance did not exceed 110 mL/min, although 51Cr-EDTA exceeded CCr by almost 30% (23). In those without SCD, the urinary clearance of 51Cr-EDTA is 85 to 95% that of CIn, whereas the clearances of technetium-99m diethylenetriamine penta-acetic acid and iodine 125 iothalamate are nearly identical to that of inulin (51). Because CCr usually exceeds CIn in healthy individuals, its validity in SCD is uncertain. The simplest estimation of GFR, by several formulas, uses plasma creatine (PCr) alone (52), but an additional problem in the use of these formulas is the greater overestimation of GFR by CCr in SCD patients than in those without SCD. The plasma clearance of 51Cr-EDTA after bolus injection approximates CIn in healthy individuals (53) but at high clearances is far less accurate (54).
For most clinical purposes, an accurate measure of GFR is unnecessary. A decrease in GFR, particularly when accompanied by proteinuria, is ominous (55). However, the effects of treatment are difficult to assess in other than a clinical research environment, although simplified methods for nonisotopic iothalamate and PAH measurement are now available (56).
Proteinuria
Proteinuria detected by dipstick in a patient with SCD should be quantified and renal function assessed. Diseases other than sickle cell glomerulopathy should be considered. If hematuria is present, RBC casts may point to pathology other than sickle cell glomerulopathy. Hypertension, hypocomplementemia, and the presence of antinuclear anti-bodies also suggest other diagnoses. Judging from the relatively uniform findings in our series (34), few other additional studies are indicated.
Treatment
The course of the progression of FSGS to chronic renal insufficiency (CRI) in SCD remains difficult to assess, in part because of the difficulties in quantitation of renal function described earlier (56). Unless the cause is known, it is difficult to prescribe a treatment for the nephropathy of SCD. The patients with proteinuria in our series (34) were not those with the most frequent sickle crises or the severest anemia. Nevertheless, some factors in the sickle cell condition must predispose to the nephropathy. Therefore it is reasonable to attempt to minimize sickling and those factors known to promote FSGS in other primary diseases or in animal models, but whether hemodynamic alterations can change the progression of FSGS to CRI is controversial (57).
Protein Restriction
A high protein intake accelerates the development of FSGS in uninephrectomized rats without necessarily causing glomerular hyperperfusion (58). It is therefore attractive to consider protein restriction in the management of SCD nephropathy, as is being tried in several forms of renal disease. In children, restriction of protein intake may carry unreasonable risks (59). Delayed growth and development is already a particular risk in the SCD patient. Therefore, we advise only the avoidance of an unusually high protein intake (greater than the recommended dietary allowance).
Angiotensin-Converting Enzyme Inhibition
Glomerular hyperperfusion and proteinuria could be mediated through increased glomerular capillary pressure, reduction of which by ACE inhibition might protect the glomerulus from FSGS.
In a 2-week trial of enalapril treatment of 10 patients with mild SCD nephropathy, blood pressure, GFR (CIn), and ERPF (PAH clearance) did not change significantly, whereas proteinuria diminished by 57%, rebounding after treatment withdrawal (34). A more recent 6-month controlled trial of enalapril therapy in 22 SCD patients with microalbuminuria showed a significant decrease in the treatment group, whereas microalbuminuria increased in the control group (60). Whether long-term ACE inhibitor therapy has a salutary effect in preventing renal insufficiency is untested.
CHRONIC RENAL INSUFFICIENCY
Clinical Features
Renal failure is one of the major organ failures that occurs in SCD, almost certainly the consequence of the progression of FSGS. Of 22 patients with nephrotic syndrome, 68% developed CRI (28). Population studies of SCD have all shown a significant incidence of CRI. Of 368 patients in one sickle cell center, 4.6% had CRI (61) associated with proteinuria and increased age. In our series of 375 patients, 6.7% had CRI (34). In a series of 785 patients from California,
P.924

33 had CRI (4.2%) (62). Overall, 5 to 18% of SCD patients develop CRI (63).
The epidemiology of renal involvement in SCD may depend on other genetic characteristics that affect the levels of fetal hemoglobin (HbF), the tendency to sickle, and other factors. Powars et al. (64) initially reported the association of β-globin gene cluster haplotypes with renal involvement: the Central African Republics (CAR) haplotypes (Bantu, Cameroon) or Benin (intermediate involvement) versus non-CAR (Senegalese and Arab-Indian) phenotypes. Guasch et al. (65) found no association with these haplotypes, but instead an association with microdeletions in the α-globin gene. Microalbuminuria was found in 22 of 76 (29%) adult SCD patients but only 13% with the microdeletions, versus 40% without those microdeletions (p <.01). Those factors that affect the interactions of RBCs with endothelium may be equally important (see later).
Treatment
Specific treatment of the patient with SCD and renal failure has been poorly explored, and the problems of CRI are only magnified by SCD. Patients with renal failure, even if it is mild, sometimes have symptomatic anemia requiring transfusion. In some of these patients, treatment with erythropoietin can variably restore hemoglobin concentrations to higher levels (66). A few patients have been treated with hydroxyurea plus erythropoietin with apparent benefit (67).
Dialysis and Transplantation
The U.S. Renal Data System reported the causes of renal failure for 255,573 patients treated from 1989 through 1993 (54). Overall, 235 were SCD patients. This is far fewer than the 1% expected from the incidence of HbSS in the total population. It is possible that physicians do not offer treatment to many SCD patients who develop CRI, assuming that the other problems are insurmountable.
Ojo et al. (68) reviewed the transplant results in SCD and found 82 patients. There was no difference in the 1-year cadaveric graft survival [SCN, 78%; other causes of end-stage renal disease (ESRD), 77%], and the multivariable adjusted 1-year risk of graft loss indicated no significant effect of SCN [relative risk (RR) = 1.39, p = .149]. However, the 3-year cadaveric graft survival tended to be lower in the SCN group (48% vs. 60%, p = .055) and their adjusted 3-year risk of graft loss was significantly higher (RR = 1.60, p = .003). There was a trend toward higher survival in the SCN transplant recipients than in their dialysis-treated, wait-listed counterparts (RR = 0.14, p = .056). In comparison to the other-ESRD group (RR = 1.00), the adjusted mortality risk in the SCN group was higher both at 1 year (RR = 2.95, p = .001) and at 3 years (RR = 2.82, p = .0001) after renal transplantation. A trend was also found toward better patient survival with renal transplantation than with dialysis in end-stage sickle cell nephropathy.
In an analysis of the United Network of Organ Sharing registry from 1987 to 1996 (69), 54 patients were found with SCD and renal failure and for whom data were available on first renal transplant outcome. Patient survival for individuals with SCD was 90% at 1 year and (of 30 patients) 75% at 3 years. The proportional risk ratio (compared to those with immunoglobulin A nephropathy) was 7.8, the highest for any condition. First-graft survival for SCD patients was 82.5% at 3 years and 54% at 3 years. The proportional risk ratio for graft survival was 1.77, but after correction for deaths of patients with functioning grafts, was only 1.06 (barely an increased risk).
We have explored (J. I. Scheinman and R. Payne, unpublished data, 2002) the U.S. Renal Data System data files (2000) for evidence of the effect of SCD on the outcome of chronic renal failure. Although the diagnostic code for sickle cell nephropathy (assigned at the time of renal failure) yielded 904 patients, further exploration of patients discovered by hospitalization codes to have SS disease yielded a total of 1656 patients—237 undergoing transplantation and 1419 not undergoing transplantation. Even after other causes of renal failure (diabetes mellitus, etc.) were eliminated, SCD patients were compared to all other African American patients in life-table survival. Without transplantation, the groups differed significantly, with vastly different numbers, and projected 10-year survival was poor for both groups: 25% for African Americans and 15% for SCD patients.
Furthermore, the 36,264 African American patients undergoing transplantation had better survival rates than the 210 SCD patients, but the life-table projected survivals were still quite close, with approximately 50% survival at 15 years (Fig. 48.3). When an age-adjusted cohort of African American patients were used as controls, the difference
P.925

statistically disappeared (p <.19). Among those who did not undergo transplantation, comparison of the African American patients with the SCD patients yielded different findings, but projected survival rates for both groups were extremely poor: 25% for African American patients and 14% for SCD patients at 10 years.
FIGURE 48.3. Life-table analysis of patient survival for end-stage renal failure patients with sickle cell disease (SCD) and age-matched African-American (AA) controls, treated with kidney transplantation or dialysis alone. All curves differ significantly due to the vastly greater numbers of AA patients (J. I. Scheinman and R. Payne, unpublished data, 2002). Tx, transplantation.
A comparison of the 153 SCD patients undergoing transplantation with those who received no transplant showed a far better survival curve: 56% versus 14% at 10 years. Survival of SCD transplanted patients was also significantly better than the 133 nontransplanted patients who were on the waiting list. This was similar to the difference between all African American patients undergoing transplantation and nontransplanted African American patients assigned to the transplant waiting list. Of 957 patients listed since 1991, essentially the cyclosporin era, only 53 SCD patients received transplants whereas 898 did not undergo transplantation; survival of the patients receiving transplants is projected to be 67% at 7 years compared with 83% for the African American cohort.
These results continue to make the case that transplantation is a better option for the SCD patient with renal failure. Results may be less satisfactory than those for other African American patients, however, and grafts have been lost due to demonstrable massive sickling events.
The problems that have occurred in SCD patients after transplantation, especially the recurrent sickle crises, could have been aggravated by the increased blood viscosity associated with a rising hematocrit. These have been treated with frequent partial exchange transfusions (70). In one early series, seven of eight patients had frequent sickle crises after transplantation (71). Renal venous thrombosis and infarction have been reported (72). In one patient with HbAS, the transplanted kidney was unfortunately removed for an apparently irreversible acute rejection that was actually intrarenal sickling (71).
The management of renal transplantation in SCD patients should then focus on potential immediate problems. It is reasonable to warm the kidney with 37°C saline, to infuse dopamine at 4 μg/kg/min during and immediately after surgery, and to provide 40% oxygen and intravenous fluids to decrease viscosity. Partial exchange transfusion may be provided at 4-week intervals (71). It is possible that before an adequate erythropoietin response occurs from the transplanted kidney, recombinant erythropoietin should be given.
Sickle cell nephropathy has been reported to recur in as little as 3 years, although other factors contributed (73). Accelerated recurrence of native disease has been noted in the transplanted kidney in other diseases, notably diabetic nephropathy (74). It is possible that the use of hydroxyurea, in itself likely to interact with other immunosuppressants, can be used to prevent this. Bone marrow transplantation can cure SCD, and the possibility of coupling it with other transplantations will undoubtedly be explored. Caution will be needed, in view of the possible increase in infection with parvovirus (75), a common agent of aplastic crises that might be especially dangerous in the transplant patient.
OTHER PROBLEMS IN SICKLE CELL DISEASE RELATED TO NEPHROLOGY
Anemia
In the presence of continued hemolysis, the patient with SCD depends on continued erythropoiesis, which determines the level of anemia. Erythropoiesis can be measured by erythron transferrin uptake (76), independent of iron or transferrin saturation. A heme protein in the kidney senses a decreased level of oxygen delivery to the kidney and stimulates erythropoietin production (77). The SCD patient with renal failure loses this normal homeostatic mechanism and becomes far more anemic. Even in the presence of a competent bone marrow and erythropoietic stimulus, adequate iron is necessary. Patients with ulcers may subtly lose iron (8).
The hemoglobin level should probably be maintained at 50 to 60% of normal in patients with SCD, in view of the serious cardiac consequences of more severe anemia (78). Although an intrinsic cardiomyopathy has been suspected in SCD, a careful autopsy study of 52 patients with a mean age of 17 years concluded that anemia alone caused heart failure (78).
Sickle Crises—Clinical Features
Sickle cell crises (1) are painful episodes of vasoocclusion, often accompanied in the second or third day by fever without documented infection. Neither abnormal blood viscosity nor the number of sickle cells absolutely predicts the frequency of crises (2). The incidence of pain crises is 0.8/yr in SS disease and 1/yr in Sβ+ thalassemia (79). With greater degrees of anemia, there is less pain, probably because of diminished blood viscosity. With increasing HbF, there are proportionally fewer painful crises, which suggests a beneficial effect of even modest increases in HbF (79).
Crises may manifest in infants as the hand-foot syndrome because of poorly developed collateral circulation. The chest syndrome is accompanied by radiologic “white-out” and is life-threatening (80). The abdominal crisis is like a “surgical abdomen” but usually without rebound tenderness. In the presence of hematuria, the origin of the pain could be assumed to be the kidney. Neuromeningitis, multiinfarct dementia, or large-vessel occlusion (stroke) can occur (81), especially in infants and children (8). A sudden anemia may be caused by sequestration crises in the spleen in the first 2 years of life; thereafter, after development of splenic fibrosis, it occurs more often in the liver. Aplastic crises arise from vasoocclusion and hemolysis without compensatory erythropoiesis, often induced by a parvovirus-like agent.
P.926

Priapism
Priapism is a specialized vasoocclusion of the penis, which was found in as many as 42% of cases in one series (1). Painful, hot, tender erection, most often on waking, lasts up to 3 hours. This can be preceded by days or weeks of “stuttering.” The engorgement is of the corpus cavernosa, usually not the spongiosum in children, with the glans flaccid (8). The pain is referred to the perineum and to the abdomen; analgesia, and often exchange transfusion, is necessary. Few patients require surgical drainage or placement of a shunt. Fibrosis may result, with consequent impotence (8).
Pathophysiology of Sickling
The clinical heterogeneity of SCD is further determined by recently described aggravators of sickling related to oxidative stress. Hemoglobin from RBCs of SCD patients has increased auto-oxidation and increased generation of superoxide and H2O2, OH-, and lipid peroxidation products (82). In a transgenic model of SCD, increased oxidant stress extends renal sickling from a more limited area in the renal medulla to a more extensive distribution to the cortex (83). The aggravating factors of sickling are likely to include the agents of microvascular inflammation and constriction, and inhibitors of relaxation. Reactive species can divert vasoactive NO from mechanisms of vascular relaxation to agents of inflammation and constriction (82). Integrins (α4β1) of the sickled RBC bind to both fibronectin (an acute-phase reactant) and vascular cell adhesion molecule-1 on endothelial cells, which is increased by inflammatory cytokines such as tumor necrosis factor-α.
Hemoxygenase-1 (HO-1) defends against and degrades heme, a lipophilic pro-oxidant, formed from the splitting of hemoglobin (83), studied in a transgenic sickle mouse. In the HO-1 “knock-out” model, administration of heme protein induces monocyte chemoattractive protein-1, mediated by nuclear factor-κB, and results in renal interstitial cell inflammation (84).
Thus increased induced oxidative stress from the sickling process precipitates acute vasoocclusive disease by mechanisms that affect the endothelium, and adhesion to it, as well as vasoconstriction.
The distorted sickle cell is partly explained by dehydration of the cell by enhanced KCl co-transport, induced by cell swelling and acidification (9). Furthermore, K+ and water efflux are enhanced by transiently increased SS cell cytosolic Ca, induced by the membrane distortion.
Acute Renal Failure
A doubling or more of PCr was found in 12 of 116 patients with SCD admitted to the hospital. It was seen most often with infections and evidence of rhabdomyolysis, and in patients with lower hemoglobin levels (mean of 6.4 vs. 8.7 g/dL) (85). Volume depletion was the most common precipitating cause. In that study patient survival rate was an 83%, with recovery of function in all patients who survived, without progression to ESRD. It is likely that nonsteroidal antiinflammatory agents are partly responsible for some episodes of acute renal failure (86), in view of the likely maintenance of GFR by PG mechanisms in SCD.
Rhabdomyolysis with acute renal failure and disseminated intravascular coagulation has been seen (rarely) in patients with sickle trait after rigorous military training (87). Similarly, there is an apparently increased risk of sudden unexplained death in patients with sickle trait (88).
Hypertension
Hypertension is unusual in SCD (2 to 6%) compared to the published incidence of 28% for the African American population in the United States at all age ranges, as found by the Cooperative Study of Sickle Cell Disease (89).
In our series (34), hypertension was rare and was found only in one patient with nephrotic-range proteinuria without renal insufficiency. The one-third incidence of hypertension in Powars et al.’s patients with renal insufficiency is unexplained (38). Nevertheless, she found a positive association between blood pressure, stroke, and increased mortality in SCD patients (90).
Infections
Infections occur in HbSS disease as a result of vascular sequestration with tissue necrosis (e.g., osteomyelitis associated with avascular necrosis, or Salmonella infection with cholelithiasis) and the immune susceptibility of the splenectomized patient (1). The organisms most commonly found are Diplococcus pneumoniae, Haemophilus influenzae, Mycoplasma, and parvovirus (8). Prophylactic penicillin therapy is essential, as for any splenectomized patient. H. influenza B vaccination is also appropriate.
Urinary infections are uncommon in SCD, except perhaps in pregnancy. Technetium 99m diphosphonate scan shows an increased renal localization in HbSS. This apparently is not caused by infection but more likely by regional stasis or tubular ischemia, which leads to peritubular extravasation (91).
GENERAL TREATMENT STRATEGIES FOR SICKLE CELL DISEASE
Institution of frequent transfusion can reduce the risk of recurrence of cerebral vasculopathy (81) by decreasing the proportion of rheologically abnormal cells. Because of the risks, transfusion should be reserved for patients at greatest risk of sickling sequelae. Most recently, Doppler ultrasonography has helped identify cerebral vessels at risk of stroke (92).
P.927

Other hemoglobins modify the tendency of HbS RBCs to sickle. HbF inhibits sickling. The agent 5-azacitidine can increase HbF levels, probably by recruitment of HbF cells, but may be a carcinogen (93). The only agent able to significantly decrease sickling by myelosuppression of HbSS cells, thus increasing HbF, is hydroxyurea. Hydroxyurea is a cytotoxic agent that inhibits the growth of erythroid burst-forming units (94). In SCD, it is the SS cells that are suppressed, which allows an increase in HbF.
Charache et al. (95) achieved 8 to 18% HbF levels that were maintained over 2 years. It has been deduced that a 20% level of HbF is likely needed to produce a definitive clinical change with regard to either hemolysis or vasoocclusion (96). A trial involving 59 patients (97) suggested that a mean HbF level of 15% could be achieved. Trials of hydroxyurea therapy in adults and children (98) have shown that it provides a safe reduction in acute episodes and allows normal growth and development (99). In addition to its effect on RBC generation, hydroxyurea induces oxidative damage to SS cells that is greater than that to AA cells, even more than to HbF (100). Antioxidants such as α-tocopherol and ascorbate are protective in vitro and may enhance the safety of hydroxyurea.
Other mechanisms that can minimize the tendency to sickling and vasoocclusion have been investigated. Reversing the cellular dehydration in SCD has been explored by blocking cation-transport channels in erythrocyte membranes, which can shift cellular cation content and cell density toward normal. Clotrimazole, an antifungal drug, reduced cellular dehydration in vitro in transgenic mice with SCD and in patients with sickle cell anemia. Magnesium salts also interfere with cation transport and cause cell rehydration (101).
Whether the cellular changes induced by clotrimazole or magnesium salts are of clinical value is not known. Combinations of hydroxyurea plus clotrimazole and erythropoietin to prevent sickling have been studied in transgenic mice.
Bone marrow transplantation can cure SCD, and the possibility of combining it with other transplantations will doubtless be investigated. Less ablative conditioning for bone marrow transplantation is being explored (102). Because of the possible increase in infection with parvovirus (75), a common agent of aplastic crisis that might be especially dangerous in the transplant patient, caution is in order.
REFERENCES
1. Weatherall DF, Cleg SB, Higs DR, et al. The hemoglobinopathies. In: Scriver CR, Beaudet AL, Sly WS, et al., eds. The metabolic basis of inherited disease, 6th ed. New York: McGraw-Hill, 1992:2293–2335.
2. Oksenhendler E, Bourbigot B, Desbazeille F, et al. Recurrent hematuria in four white patients with sickle cell trait. J Urol 1984;132:1201–1203.
3. Osegbe DN. Haematuria and sickle cell disease. A report of 12 cases and review of the literature. Trop Geograph Med 1990;42:22–27.
4. Statius van Eps LAW, de Jong PE. Sickle cell disease. In: Scherer WR, Gottschalk KW, eds. Diseases of the kidney, 4th ed. Boston: Little, Brown, 1988:2561–2681.
5. Davies SC, Dewitt PE. Sickle cell disease. Br J Hosp Med 1984;31:440–444.
6. Audit JC, Ugbodaga CI, Okafor LA, et al. Urographic changes in homozygous sickle cell disease. Diagn Imag 1983;52:259–263.
7. Pandya KK, Koshy M, Brown N, et al. Renal papillary necrosis in sickle cell hemoglobinopathies. J Urol 1976;115: 497–501.
8. Chauhan PM, Kondlapoodi P, Natta CL. Pathology of sickle cell disorders. Pathol Ann 1983;18:253–276.
9. Bunn HF. Mechanisms of disease—pathogenesis and treatment of sickle cell disease. N Engl J Med 1997;337:762–769.
10. de Jong PE, Statius van Eps LW. Sickle cell nephropathy: new insights into its pathophysiology. Kidney Int 1985;27: 711–717.
11. Burry A, Cross R, Axelsen R. Analgesic nephropathy and the renal concentrating mechanism. Pathol Ann 1977;12:1–31.
12. Avery RA, Harris JE, Davis CJ, et al. Renal medullary carcinoma: clinical and therapeutic aspects of a newly described tumor. Cancer 1996;78:128–132.
13. Schultz PK, Strife JL, Strife CF, et al. Hyperechoic renal medullary pyramids in infants and children. Radiology 1991;181: 163–167.
14. Walker TM, Serjeant GR. Increased renal reflectivity in sickle cell disease: prevalence and characteristics. Clin Radiol 1995;50:566–569.
15. McCall IW, Moule N, Desai P, et al. Urographic findings in homozygous sickle cell disease. Radiology 1978;126:99–104.
16. Mapp E, Karasick S, Pollack H, et al. Uroradiological manifestations of S-hemoglobinopathy. Semin Roentgenol 1987; 22:186–194.
17. Braden GL, Kozinn DR, Hampf FE, et al. Ultrasound diagnosis of early renal papillary necrosis. J Ultrasound Med 1991;10:401–403.
18. Eknoyan G, Qunibi WY, Grissom RT, et al. Renal papillary necrosis: an update. Medicine 1982;61:55–73.
19. Sabatini S. Pathophysiologic mechanisms of abnormal collecting duct function. Semin Nephrol 1989;9:179–202.
20. Keeler R, Wilson N. Natriuretic response to hypervolemia is absent in rats with papillary necrosis. Am J Physiol 1989;257: R422–R426.
21. Allon M, Lawson L, Eckman JR, et al. Effects of nonsteroidal antiinflammatory drugs on renal function in sickle cell anemia. Kidney Int 1988;34:500–506.
22. Beutler E. Erythrocyte disorders: anemias related to abnormal globulin. In: Williams WJ, Beutler E, Erslev AJ, et al., eds. Hematology, 4th ed. New York: McGraw-Hill, 1992:613–626.
23. Allon M. Renal abnormalities in sickle cell disease. Arch Intern Med 1990;150:501–504.
24. Buckalew VM, Someren A. Renal manifestations of sickle cell disease. Arch Intern Med 1974;133:660–669.
P.928

25. Devereux S, Knowles SM. Rhabdomyolysis and acute renal failure in sickle cell anaemia. BMJ 1985;290:1707.
26. de Jong PE, de Jong-van den Berg LT, Schouten H, et al. The influence of indomethacin on renal acidification in normal subjects and in patients with sickle cell anemia. Clin Nephrol 1983;19:259–264.
27. Kurtzman NA. Acquired distal renal tubular acidosis. Kidney Int 1983;24:807–819.
28. Bakir AA, Hathiwala SC, Ainis H, et al. Prognosis of the nephrotic syndrome in sickle glomerulopathy. A retrospective study. Am J Nephrol 1987;7:110–115.
29. DeFronzo RA, Taufield PA, Black H, et al. Impaired renal tubular potassium secretion in sickle cell disease. Ann Intern Med 1979;90:310–316.
30. Morgan AG, de Ceulaer K, Serjeant GR. Glomerular function and hyperuricemia in sickle cell disease. J Clin Pathol 1984;37:1046–1049.
31. de Jong PE, de Jong-van den Berg LT, Donker AJ, et al. The role of prostaglandins and renin in sickle-cell nephropathy. A hypothesis. Neth J Med 1978;21:67–72.
32. Berman LB, Tublin I. The nephropathies of sickle-cell disease. Arch Intern Med 1959;103:602–606.
33. Sklar AH, Campbell H, Caruana RJ, et al. A population study of renal function in sickle cell anemia. Int J Artif Organs 1990;13:231–236.
34. Falk RJ, Scheinman JI, Phillips G, et al. Prevalence and pathologic features of sickle cell nephropathy and response to inhibition of angiotensin-converting enzyme. N Engl J Med 1992;326:910–915.
35. Guasch A, Cua M, Mitch WE. Early detection and the course of glomerular injury in patients with sickle cell anemia. Kidney Int 1996;49:786–791.
36. Guasch A, Cua M, You W, et al. Sickle cell anemia causes a distinct pattern of glomerular dysfunction. Kidney Int 1997;51: 826–833.
37. Lonsdorfer A, Comoe L, Yapo AE, et al. Proteinuria in sickle cell trait and disease: an electrophoretic. Clin Chim Acta 1989;181:239–248.
38. Powars DR, Elliott-Mills DD, Chan L, et al. Chronic renal failure in sickle cell disease: risk factors, clinical course, and mortality. Ann Intern Med 1991;115:614–620.
39. Zamurovic D, Churg J. IDiopathic and secondary mesangiocapillary glomerulonephritis. Nephron 1984;38:145–153.
40. Morgan AG, Shah DJ, Williams W. Renal pathology in adults over 40 with sickle-cell disease. West Indian Med J 1987;36:241–250.
41. Bhathena DB, Sondheimer JH. The glomerulopathy of homozygous sickle hemoglobin (SS) disease: morphology and pathogenesis. J Am Soc Nephrol 1991;1:1241–1252.
42. de Jong PE, Saleh AW, de Zeeuw D, et al. Urinary prostaglandins in sickle cell nephropathy: a defect in 9-ketoreductase activity? Clin Nephrol 1984;22:212–213.
43. Strauss J, Pardo V, Koss MN, et al. Nephropathy associated with sickle cell anemia: an autologous immune complex nephritis. I. Studies on nature of glomerular-bound antibody and antigen identification in a patient with sickle cell disease and immune deposit glomerulonephritis. Am J Med 1975;58:382–387.
44. Jennette JC, Charles L, Grubb W. Glomerulomegaly and focal segmental glomerulosclerosis associated with obesity and sleep-apnea syndrome. Am J Kidney Dis 1987;10:470–472.
45. Olson JL, Hostetter TH, Rennke HG, et al. Altered glomerular permselectivity and progressive sclerosis following extreme ablation of renal mass. Kidney Int 1982;22:112–126.
46. Chen YT, Coleman RA, Scheinman JI, et al. Renal disease in type I glycogen storage disease. N Engl J Med 1988;318:7–11.
47. Spear G. Glomerular alterations in cyanotic congenital heart disease. Bull Johns Hopkins Hosp 1960;106:347–367.
48. Lande IM, Glazer GM, Sarnaik S, et al. Sickle-cell nephropathy: MR imaging. Radiology 1986;158:379–383.
49. Yoshida Y, Fogo A, Ichikawa I. Glomerular hemodynamic changes vs. hypertrophy in experimental glomerular sclerosis. Kidney Int 1989;35:654–660.
50. Hostetter TH, Olson JL, Rennke HG, et al. Hyperfiltration in remnant nephrons: a potentially adverse response to renal ablation. Nephron 1981;241:F85-F93.
51. Haycock G. Creatinine, body size and renal function. Pediatr Nephrol 1989;3:22–24.
52. Schwartz GJ, Haycock GB, Edelmann CM, et al. A simple estimate of glomerular filtration rate in children derived from body length and plasma creatinine. Pediatrics 1976;58:259–263.
53. Mulhern JG, Perrone RD. Accurate measurement of glomerular filtration rate. Int Yearb Nephrol 1990;277–291.
54. Excerpts from the United States Renal Data System 1996 Annual Data Report. Am J Kidney Dis 1996;28:S1–165.
55. Walser M. Progression of chronic renal failure in man. Kidney Int 1990;37:1195–1210.
56. Coakley DF, Roland CL, Falk RJ, et al. Renal function assessment with HPLC analysis of iothalamate (IOTH) and para-aminohippurate (PAH) compared to inulin, PAH, and creatinine as measured by standard analytical methods. Pharmacotherapy 1991;11:265.
57. Woolf AS, Fine LG. Do glomerular hemodynamic adaptations influence the progression of human renal disease? Pediatr Nephrol 1991;5:88–93.
58. Morozumi K, Thiel G, Gudat F, et al. Studies on morphological outcome of cyclosporine-associated arteriolopathy after discontinuation of cyclosporine in patients with renal allografts. Transplant Proc 1993;25:537–539.
59. Raymond NG, Dwyer JT, Nevins P, et al. An approach to protein restriction in children with renal insufficiency. Pediatr Nephrol 1990;4:145–151.
60. Foucan L, Bourhis V, Bangou J, et al. A randomized trial of captopril for microalbuminuria in normotensive adults with sickle cell anemia. Am J Med 1998;104:339–342.
61. Heeg JE, de Jong PE, van der Hem GK, et al. Reduction of proteinuria by angiotensin converting enzyme inhibition. Kidney Int 1987;32:78–83.
62. Murthy VS, Haywood J. Survival analysis by sex, age, group and hemotype in sickle cell disease. J Chronic Dis 1981;34:3 13–319.
63. Thomas AN, Pattison C, Serjeant GR. Causes of death in sickle-cell disease in Jamaica. BMJ 1982;285:633–635.
64. Powars DR, Meiselman HJ, Fisher TC, et al. Beta-S gene cluster haplotypes modulate hematologic and hemorheologic expression in sickle cell anemia. Use in predicting clinical severity. Am J Pediatr Hematol Oncol 1994;16:55–61.
P.929

65. Guasch A, Zayas CF, Eckman JR, et al. Evidence that microdeletions in the alpha globin gene protect against the development of sickle cell glomerulopathy in humans. J Am Soc Nephrol 1999;10:1014–1019.
66. Tomson CR, Edmunds ME, Chambers K, et al. Effect of recombinant human erythropoietin on erythropoiesis in homozygous sickle-cell anaemia and renal failure. Nephrol Dial Transplant 1992;7:817–821.
67. Rodgers GP, Dover GJ, Uyesaka N, et al. Augmentation by erythropoietin of the fetal-hemoglobin response to hydroxyurea in sickle cell disease [see comments]. N Engl J Med 1993;328:73–80.
68. Ojo AO, Govaerts TC, Schmouder RL, et al. Renal transplantation in end-stage sickle cell nephropathy. Transplantation 1999;67:291–295.
69. Bleyer AJ, Donaldson LA, McIntosh M, et al. Relationship between underlying renal disease and renal transplantation outcome. Am J Kidney Dis 2001;37:1152–1161.
70. Spector D, Zachary JB, Sterioff S, et al. Painful crises following renal transplantation in sickle cell anemia. Am J Med 1978;64:835–839.
71. Chatterjee SN, Lundberg GD, Berne TV. Sickle cell trait: possible contributory cause of renal allograft failure. Urology 1978;11:266–268.
72. Donnelly PK, Edmunds ME, O’Reilly K. Renal transplantation in sickle cell disease [Letter]. Lancet 1988;2:229.
73. Miner DJ, Jorkasky DK, Perloff LJ, et al. Recurrent sickle cell nephropathy in a transplanted kidney. Am J Kidney Dis 1987;10:306–313.
74. Mauer SM, Steffes MW, Connett J, et al. The development of lesions in the glomerular basement membrane and mesangium after transplantation of normal kidneys to diabetic patients. Diabetes 1983;32:948–952.
75. Khamashta MA, Williams FM, Hunt BJ. Anticoagulation for venous thromboembolism [Letter]. N Engl J Med 1999; 341:539–540.
76. Cazzola M, Pootrakul P, Huebers HA, et al. Erythroid marrow function in anemic patients. Blood 1987;69:296–301.
77. Goldberg JS, Dunning SP, Bunn HF. Regulation of the erythropoietin gene: evidence that the oxygen sensor is a heme protein. Science 1988;242:1412–1415.
78. Sztajzel J, Ruedin P, Stoermann C, et al. Effects of dialysate composition during hemodialysis on left ventricular function. Kidney Int 1993;41:S60–S66.
79. Platt OS, Thorington BD, Brambilla DJ, et al. Pain in sickle cell disease. Rates and risk factors. N Engl J Med 1991;325: 11–16.
80. Ponez M, Kane E, Gill FM. Acute chest syndrome in sickle cell disease: etiology and clinical correlates. J Pediatr 1985; 107:861–866.
81. Powars DR. Sickle cell anemia and major organ failure. Hemoglobin 1990;14:573–598.
82. Aslan M, Thornley-Brown D, Freeman BA. Reactive species in sickle cell disease. Ann N Y Acad Sci 2000;899:375–391.
83. Nath KA, Grande JP, Haggard JJ, et al. Oxidative stress and induction of heme oxygenase-1 in the kidney in sickle cell disease. Am J Pathol 2001;158:893–903.
84. Nath KA, Vercellotti GM, Grande JP, et al. Heme protein-induced chronic renal inflammation: suppressive effect of induced heme oxygenase-1. Kidney Int 2001;59:106–117.
85. Sklar AH, Perez JC, Harp RJ, et al. Acute renal failure in sickle cell anemia. Int J Artif Organs 1990;13:347–351.
86. Simckes AM, Chen SS, Osorio AV, et al. Ketorolac-induced irreversible renal failure in sickle cell disease: a case report. Pediatr Nephrol 1999;13:63–67.
87. Koppes GM, Daly JJ, Coltman CA, et al. Exertion-induced rhabdomyolysis with acute renal failure and disseminated intravascular coagulation in sickle cell trait. Am J Med 1977;63:313–317.
88. Kark JA, Posey DM, Schumacher HR, et al. Sickle-cell trait as a risk factor for sudden death in physical training. N Engl J Med 1987;317:781–787.
89. Johnson CS, Giorgio AJ. Arterial blood pressure in adults with sickle cell disease. Arch Intern Med 1981;141:891–893.
90. Pegelow CH, Colangelo L, Steinberg M, et al. 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 1997;102:171–177.
91. Sty JR, Babbitt DP, Sheth K. Abnormal Tc-99m-methylene diphosphonate accumulation in the kidneys of children with sickle cell disease. Clin Nucl Med 1980;5:445–447.
92. Adams RA, McKie V, Nichols F, et al. The use of transcranial ultrasonography to predict stroke in sickle cell disease. N Engl J Med 1992;326:605–610.
93. Shannon KM. Recombinant erythropoietin in pediatrics: a clinical perspective. Pediatr Ann 1990;19:197–206.
94. Baliga BS, Pace BS, Chen HH, et al. Mechanism for fetal hemoglobin induction by hydroxyurea in sickle cell erythroid progenitors. Am J Hematol 2000;65:227–233.
95. Charache S, Dover GJ, Moyer MA, et al. Hydroxyurea-induced augmentation of fetal hemoglobin production in patients with sickle cell anemia. Blood 1987;69:109–116.
96. Noguchi CT, Rodgers GP, Serjeant G, et al. Current concepts: levels of fetal hemoglobin necessary for treatment of sickle cell disease. N Engl J Med 1988;318:96–99.
97. Charache S, Dover GJ, Moore RD, et al. Hydroxyurea: effects on hemoglobin F production in patients with sickle cell anemia. Blood 1992;79:2555–2565.
98. Ferster A, Tahriri P, Vermylen C, et al. Five years of experience with hydroxyurea in children and young adults with sickle cell disease. Blood 2001;97:3628–3632.
99. Wang WC, Helms RW, Lynn HS, et al. Effect of hydroxyurea on growth in children with sickle cell anemia: results of the HUG-KIDS Study. J Pediatr 2002;140:225–229.
100. Iyamu EW, Fasold H, Roa D, et al. Hydroxyurea-induced oxidative damage of normal and sickle cell hemoglobins in vitro: amelioration by radical scavengers. J Clin Lab Anal 2001;15:1–7.
101. Steinberg MH. Management of sickle cell disease. N Engl J Med 1999;340:1021–1030.
102. Schleuning M, Stoetzer O, Waterhouse C, et al. Hematopoietic stem cell transplantation after reduced-intensity conditioning as treatment of sickle cell disease. Exp Hematol 2002; 30:7–10.