ORIGINAL ARTICLE
Prevalence and Progression of Chronic Kidney Disease in Adult Patients With Sickle Cell Disease Elvira O. Gosmanova, MD, FASN,* Sahar Zaidi, MD,Þ Jim Y. Wan, PhD,þ and Patricia E. Adams-Graves, MDÞ
Aim: We evaluated the prevalence and progression of chronic kidney disease (CKD) during the 5-year period in a cohort of patients with sickle cell disease (SCD) aged 18 years and older. Methods: We studied 98 patients with SCD. Chronic kidney disease stages I through V were defined based on estimated glomerular filtration rate (eGFR), and albuminuria grades were defined based on spot urine protein-to-creatinine ratio according to the 2012 Kidney Disease Improving Global Outcomes recommendations. In patients with eGFR of greater than 60 mL/min per 1.73 m2, CKD was diagnosed if grade A2 or A3 albuminuria was present. Chronic kidney disease progression was defined as an increase in CKD stage with an additional eGFR reduction of more than 25% from baseline. Results: At baseline, 28.6% of patients had CKD. After a mean follow-up of 5.0 (SD, 0.9) years, 17 patients developed new CKD and the overall CKD prevalence increased to 41.8%. In addition, 8 patients experienced CKD progression. The following baseline variables were associated with the development and progression of CKD in univariate analysis: older age (P = 0.003), higher systolic blood pressure (BP; P = 0.003), lower eGFR (P = 0.001), higher serum creatinine (P = 0.001), and A3 albuminuria (P = 0.008). In multivariate analysis, baseline A3 albuminuria (adjusted odds ratio, 5.0; 95% confidence interval, 1.1-24.3; P = 0.048) and each 1-mm Hg increase in systolic BP (adjusted odds ratio, 1.04; 95% confidence interval, 1.0-1.07; P = 0.039) predicted CKD development and progression. Conclusions: Chronic kidney disease is common in patients with SCD and its prevalence increases with age. Several baseline modifiable and nonmodifiable factors were associated with the development and progression of CKD in patients with SCD. Strategies targeting BP control and proteinuria may be beneficial for individuals with SCD. Key Words: chronic kidney disease, sickle cell disease (J Investig Med 2014;62: 804Y807)
S
ickle cell disease (SCD) is a group of genetic disorders in which a A-globin chain of hemoglobin is affected by several mutations (eg, hemoglobin S, C, A-thalassemia).1 End-organ damage is the major cause of morbidity and mortality in adult patients with SCD.2,3 The survival of patients with SCD is significantly reduced when chronic kidney disease (CKD) is also
From the Divisions of *Nephrology, and †Hematology, Department of Medicine, and ‡Department of Preventive Medicine, University of Tennessee Health Science Center, Memphis, TN. Received November 18, 2013, and in revised form March 3, 2014. Accepted for publication March 14, 2014. Reprints: Elvira O. Gosmanova, MD, FASN, 956 Court Ave, Suite B226, Memphis, TN 38103. E-mail: [email protected]. Disclaimers: None of the authors who participated in the preparation of the article have any potential conflicts of interest. Source of funding: None. This work was not supported by any grant or funding. Copyright * 2014 by The American Federation for Medical Research ISSN: 1081-5589 DOI: 10.1097/01.JIM.0000446836.75352.72
804
present.4,5 Although sickle cell nephropathy is a rare cause of end-stage kidney disease, the survival of these patients is also significantly lower than that of patients with end-stage kidney disease from other etiologies.6 The prevalence of CKD in patients with SCD is dependent on the definitions used to describe the kidney dysfunction. Around 50% of patients with SCD are reported to have estimated glomerular filtration rates (eGFRs) in the hyperfiltration range,7 and 21% to 27% of patients with SCD have eGFR of less than 90 mL/min per 1.73 m2.8 In addition, up to 68% of patients with hemoglobin SS disease and 42% of patients with other sickling hemoglobinopathies were found to have abnormal urinary albumin excretion (in the microalbuminuric or macroalbuminuric range).8 However, to our knowledge, no previous studies have assessed the prevalence of CKD based on the recent Kidney Disease Improving Global Outcomes (KDIGO) recommendations that incorporate eGFR and levels of albuminuria.9 The aim of this retrospective analysis was to evaluate the prevalence and progression of CKD during the 5-year period in a cohort of patients with SCD aged 18 years and older.
METHODS Study Design and Patient Population This was a single-center retrospective cohort study conducted by means of a review of medical records at the Diggs-Kraus Comprehensive Sickle Cell Center, which is affiliated with the University of Tennessee Health Science Center and located at the Regional Medical Center in Memphis, Tenn. We identified patients with SCD who were evaluated at the clinic between January and December 2006 and for whom demographic and clinical data of interest were available from both baseline and follow-up examinations. Individuals with the sickle cell trait were not included. Baseline data collected from the medical records included the patient’s age, sex, race, variant of sickle cell hemoglobinopathy (sickle SS disease and other hemoglobinopathies including sickle SC disease and sickle S-A-thalassemia), serum creatinine (SCr) concentration, serum hemoglobin and total bilirubin levels, urinary protein excretion (UPE), systolic and diastolic blood pressure (BP), hydroxyurea (HU) use, and status with regard to chronic hepatitis B and C as well as human immunodeficiency virus. The follow-up data collected closest to December 2011 included data on SCr concentration and UPE. The final cohort for analysis included 98 patients with SCD who were selected based on the availability of all of the necessary information from both baseline and follow-up. Urinary protein was measured using a colorimetric pyrocatechol violet dye-binding assay (Vitros; Ortho-Clinical Diagnostics, Rochester, NY) and urine creatinine using an enzymatic technique (Vitros; Ortho-Clinical Diagnostics). Urinary protein excretion, expressed as the urine protein-tocreatinine ratio (UPCR) in milligram per gram creatinine, was defined according to the 2012 KDIGO clinical practice guidelines9 as grade A1 or normal albuminuria (UPCR, G150 mg/g), A2 or microalbuminuria (UPCR, between 150-500 mg/g), or A3 Journal of Investigative Medicine
&
Volume 62, Number 5, June 2014
Copyright © 2014 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
Journal of Investigative Medicine
&
Volume 62, Number 5, June 2014
FIGURE 1. Prevalence of CKD stages in patients with SCD.
or macroalbuminuria (UPCR, 9500 mg/g). Kidney function was assessed by estimating glomerular filtration rate (eGFR) as calculated according to the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.10 We evaluated the prevalence of each CKD stage at baseline and after 5 years of follow-up. Chronic kidney disease stages were defined as follows: stage I, eGFR of greater than or equal to 90 mL/min per square meter with the presence of abnormal albuminuria (grade A2 or A3); stage II, eGFR of 60 to 89 mL/min per 1.73 square meter with the presence of abnormal albuminuria; stage III, eGFR of 30 to 59 mL/min per 1.73 square meter; stage IV, eGFR of 15 to 29 mL/min per square meter; and stage V, eGFR of less than 15 mL/min per 1.73 square meter. Hyperfiltration was defined as eGFR of greater than 130 mL/min per 1.73 m2 for women and greater than 140 mL/min per 1.73 m2 for men.11 Chronic kidney disease progression was defined as an increase in CKD stage with an additional 25% reduction in eGFR from baseline.9
Statistical Analysis Statistical analysis was performed using SAS 9.2 (SAS Institute Inc, Cary, NC). All associations with P of less than 0.05 were considered significant. Baseline characteristics were compared using the Mann-Whitney U test and the Fisher exact probability test for continuous and categorical data, respectively. The association between baseline variables and combined outcome with regard to the development and progression of CKD was tested using multivariate regression analysis to model the odds ratios (ORs) with 95% confidence interval (CI) of the cumulative probability of new and worsening CKD among patients with SCD with different baseline characteristics. The following baseline covariates were investigated: age, sex, hemoglobin SS disease versus other hemoglobinopathies, systolic and diastolic BP, baseline SCr, eGFR, baseline serum hemoglobin and total bilirubin concentration, and albuminuria grade. Hepatitis B, hepatitis C, and human immunodeficiency virus were present in 3 patients, 5 patients, and 1 patient, respectively; these patients were not included in multivariate analysis.
RESULTS Among the 98 patients with SCD included in the final analysis, the mean (SD) age for the whole cohort was 31.6 (10.6) years; 43% were men, and 56% had hemoglobin SS disease, whereas 44% had other hemoglobinopathies. At baseline, the mean (SD) SCr was 0.79 (0.42) mg/dL (66.9 [36.9] Kmol/L), the mean (SD) eGFR was 126.0 (32.7) mL/min per 1.73 m2, and 17.3% and 9.2% of patients had grade A2 and A3 albuminuria, respectively. The mean (SD) BP was 120.2/74.0 (15.8/11.8) mm Hg, the mean (SD) hemoglobin was 9.5 (1.9) g/dL, the mean (SD) serum total bilirubin was 3.1 (2.5) mg/dL, and 22% of patients were using HU.
CKD in Patients With Sickle Cell Disease
At baseline, 28 (28.6%) patients met the criteria for CKD diagnosis. The distribution of the various CKD stages is shown in Figure 1. Hyperfiltration was present in 47% of patients. The presence of CKD at baseline was associated with higher SCr (P = 0.001), lower eGFR (P = 0.002), presence of baseline grade A2 or A3 albuminuria (P G 0.001), and lower baseline hemoglobin (P = 0.042; Table 1). The type of hemoglobinopathy, age, sex, baseline total bilirubin concentration, levels of systolic and diastolic BP, and the use of HU were not associated with the presence of CKD (Table 1). After a mean (SD) follow-up of 5 (0.9) years, 17 patients had developed new CKD (defined in all but 1 patient as development of A2 or A3 albuminuria), and the overall prevalence of CKD had increased to 41.8%. The prevalence of stage I CKD had increased from 19.4% to 28.6%, that of stage II CKD had increased from 4.1% to 8.2%, and that of stage IV CKD had increased from 1% to 2%, whereas that of stage III CKD had decreased from 4.1% to 1%. During this time, 2% of patients progressed to CKD stage V (Fig. 1). At the end of the follow-up period, 65.4% of patients with abnormal albuminuria at baseline had retained stable UPE, whereas 11.5% of patients had progressed from grade A2 albuminuria to grade A3, and 23.1% of patients had regressed from a higher grade of albuminuria to a lower grade (2/3 of these from A2 to A1 and 1/3 of these from A3 to A2). Among the patients with normal albuminuria at baseline, 17 (17.3%) had developed abnormal albumin excretion (10 patients developed grade A2 albuminuria, and 7 patients developed grade A3). In unadjusted analysis, older age (P = 0.003), higher baseline systolic BP (P = 0.003), higher baseline SCr (P = 0.001), lower eGFR (P = 0.001), and A3 albuminuria (P = 0.008) were associated with combined outcome of CKD development or progression (Table 2). In multivariate analysis that excluded albuminuria, each decrease of 10 mL/min per 1.73 m2 in eGFR was associated with increased risk of CKD development or progression (adjusted OR, 1.49; 95% CI, 1.10-2.01; P = 0.01). In adjusted analysis that excluded eGFR, the presence of A3
TABLE 1. Baseline Characteristics Associated With the Presence of CKD in Patients With SCD CKD Patients Baseline Patient Characteristics Age, mean (SD), y Sex, male, % Race, African American, % Hemoglobin SS disease, % SCr, mean (SD), mg/dL eGFR, mean (SD), mL/min per 1.73 m2 UPCR, 90.15 mg/g, % Serum hemoglobin, mean (SD), g/dL Total bilirubin, mean (SD), mg/dL SBP, mean (SD), mm Hg DBP, mean (SD), mm Hg HU use, %
Yes (n = 28)
No (n = 70)
P
33.8 (11.9) 57.1 96.4 60.7 1.0 (0.66) 112.3 (43.2)
30.5 (9.6) 38.6 97.1 52.9 0.7 (0.2) 133.8 (23.1)
NS NS NS NS 0.001 0.002
92.8 9.1 (1.9)
0 9.8 (1.7)
G0.001 0.042
3.7 (2.7)
2.9 (2.4)
NS
124.5 (12.6) 76.8 (12.6) 32.1
118.6 (16.2) 73.0 (11.5) 21.4
NS NS NS
DBP indicates diastolic BP; NS, not significant; SBP, systolic BP.
* 2014 The American Federation for Medical Research
Copyright © 2014 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
805
Journal of Investigative Medicine
Gosmanova et al
TABLE 2. Risk Factors Associated With CKD Development or Progression in Patients With SCD CKD Development or Progression Patients Baseline Patient Characteristics
Yes (n = 25)
No (n = 73)
Age, mean (SD), y 36.7 (11.1) 29.7 (9.5) Sex, male, % 52.0 41.2 Hemoglobin SS disease, % 56.0 51.2 SCr, mean (SD), mg/dL 1.0 (0.7) 0.7 (0.2) 109.5 (42.5) 133.9 (24.2) eGFR, mean (SD), mL/min per 1.73 m2 UPCR, 90.5 mg/g, % 24 4.1 Serum hemoglobin, 9.6 (2.1) 9.6 (1.6) mean (SD), g/dL Total bilirubin, mean (SD), g/dL 3.3 (3.0) 3.0 (2.3) SBP, mean (SD), mm Hg 128.4 (19.9) 117.6 (11.4) DBP, mean (SD), mm Hg 78.0 (13.0) 72.8 (11.3) HU use, % 28.0 21.9
P 0.003 0.3 0.9 0.001 0.001 0.008 0.9 0.7 0.003 0.062 0.5
DBP indicates diastolic BP; SBP, systolic BP.
albuminuria (OR, 5.0; 95% CI, 1.1Y24.3; P = 0.048) was predictive of CKD development and progression. In addition, each 1-mm Hg increase in systolic BP was associated with an OR of 1.04 (95% CI, 1.01Y1.07; P = 0.039) for combined outcome of the development or progression of CKD. The associations between systolic BP and abnormal albuminuria and CKD development or progression were attenuated by adjustment for eGFR (P 9 0.1), suggesting that eGFR may be an intermediate variable among these risk factors and the combined renal outcome of CKD development and progression.
DISCUSSION The concomitant presence of CKD adversely affects the health outcomes of patients with SCD. Patients with SCD and elevated SCr have reduced survival compared with patients with SCD with no nephropathy.4,5 Therefore, it is important to understand the prevalence and progression of CKD in these patients. To our knowledge, this is the first study to evaluate the prevalence of various CKD stages in patients with SCD based on the KDIGO guidelines combining eGFR and levels of albuminuria.9 It has been reported that 4% to 7% of patients with SCD have elevated SCr.12,13 As in a study from Brazil, we found that 5.1% of patients with SCD had an eGFR of less than 60 mL/min per 1.73 m2 using the CKD-EPI equation.7 Although the overall number of patients with moderate to severe CKD was small, 80% of them (4/5 patients) progressed to a more advanced CKD stage during the 5-year period. Another study used eGFR of less than 90 mL/min per 1.73 m2, a value derived from the CockroftGault equation, as the threshold for CKD, and it found CKD to be present in 21% of patients with SCD, an incidence that was higher than the 13% seen in the current study, applying a similar eGFR threshold.8 As recommended by the KDIGO, we used the CKD-EPI equation as the preferred method of estimating GFR in the absence of data, suggesting that other equations would be superior for estimating GFR in patients with SCD.9 The CKDEPI equation has been shown to be more accurate in the estimation of true GFR with values above 60 mL/min per 1.73 m2.10
806
&
Volume 62, Number 5, June 2014
Similar to previous reports,7,11 we found that hyperfiltration was present in 47% of patients with SCD. We observed that lower hemoglobin concentration was associated with the presence of hyperfiltration (data not shown). A link was recently proposed between hemolysis that results in lower serum hemoglobin concentrations in patients with SCD and hyperfiltration.14 The heme released during hemolysis of sickling red blood cells can activate an inducible hemoxygenase and increase the formation of carbon monoxide, which, in turn, causes renal vasodilatation, increased intraglomerular pressure, and subsequent hyperfiltration.14 A total of 20.4% of patients with SCD with eGFR in the hyperfiltration range had increased albuminuria (A2 or A3). These data suggest that a large number of patients with normal and elevated GFR meet the KDIGO definition for the presence of stage I CKD. Several relatively small and short-term studies have evaluated the antiproteinuric effect of angiotensin converting enzyme inhibitors in patients with SCD12,15 and have shown that angiotensin converting enzyme inhibitors effectively reduce urinary albumin excretion; nevertheless, long-term studies are needed to demonstrate whether this reduction in albuminuria can also result in a slowing of CKD progression in patients with SCD. Previously, a high incidence (68%) of abnormal albuminuria, determined from the urine albumin-to-creatinine ratio, was reported in patients with SCD.8 In the present study, the prevalence of abnormal albuminuria at baseline was lower at 26.5% but subsequently increased to 42.8% after 5 years of follow-up. There is an important methodological difference between the albuminuria measurement techniques used in the study by Guasch et al.8 and the current study. The former study used the albumin-to-creatinine ratio for albuminuria classification. We used the UPCR to determine the albuminuria grades (A1-normal, A2, or microalbuminuria, and A3, or macroalbuminuria) based on the 2012 KDIGO recommendations, which support UPCR as an acceptable alternative method for albuminuria gradation when direct urine albumin measurements are not available.9 We were also able to analyze the natural history of albuminuria progression. We observed that during the 5-year follow-up period, most of the patients with abnormal albuminuria at baseline (65.4%) had unchanged UPE, whereas a smaller number (11.5%) of patients progressed to a higher albuminuria grade, and 23.1% of patients regressed to a lower albuminuria grade (most [2/3] from A2 to normal but some [1/3] from A3 to A2 albuminuria). After 5 years of follow-up, 17.3% of patients developed new albuminuria. It has previously been shown that older age and higher systolic BP are associated with the presence of albuminuria.8 We also found that older age and higher baseline systolic BP were associated with the separate outcome of new albuminuria development (data not shown). Age is not a modifiable factor, but careful monitoring and treatment of elevated BP to reduce the likelihood of developing CKD are potentially achievable goals. The difficulty lays in determining optimal BP values, above which BP should be treated, for patients with SCD. The average systolic and diastolic BP in patients with SCD are similar to those in the general population in individuals aged 18 to 45 years; in individuals aged older than 45, however, systolic and diastolic BP tend to be lower in patients with SCD than in similar individuals without SCD.8 Forty-four percent and ten percent of patients with SCD are reported to have systolic and diastolic BP between 120 to 140/80 to 90 mm Hg and greater than 140/90 mm Hg, respectively.16 Despite the fact that most of the patients have BP values below the hypertensive range, BP values in the prehypertension range (120Y140/ 80Y90 mm Hg) have been shown to be adversely associated with * 2014 The American Federation for Medical Research
Copyright © 2014 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
Journal of Investigative Medicine
&
Volume 62, Number 5, June 2014
the presence of pulmonary hypertension and elevated SCr16; this suggests that BP levels greater than 120/80 mm Hg might be considered ‘‘relative’’ hypertension in patients with SCD. We also found that even small increases in systolic BP were associated with increased risk of CKD development and progression. Before routine antihypertensive treatment can be instituted for BP of greater than 120/80 mm Hg, however, this hypothesis requires further testing in prospective trials. Several important limitations of the present study need to be recognized. This was a single-site observation study with a relatively small study population. Not all potential confounders that may have influenced the development of renal dysfunction in patients with SCD were investigated; one notable confounder that was left out is >-thalassemia status, which may be protective against kidney dysfunction. We also used UPCR to characterize albuminuria, whereas urine albumin to creatinine ratio is more sensitive in the detection of low levels of urinary albumin (G300 mg/d). The main strength of this study is that it is the first to provide information about the prevalence of the various CKD stages and their progression in patients with SCD. In conclusion, CKD in patients with SCD is common and its prevalence increases with age. Chronic kidney disease was observed in 35.3% of patients with SCD aged younger than 40 years old and in 53.3% of patients with SCD aged older than 40. Strategies targeting control of BP and proteinuria may be beneficial for individuals with SCD. REFERENCES 1. Kato GJ, Hebbel RP, Steinberg MH, et al. Vasculopathy in sickle cell disease: Biology, pathophysiology, genetics, translational medicine, and new research directions. Am J Hematol. 2009;84(9):618Y625. 2. Powars DR, Chan LS, Hiti A, et al. Outcome of sickle cell anemia: a 4-decade observational study of 1056 patients. Medicine (Baltimore). 2005;84(6):363Y376. 3. Serjeant GR, Serjeant BE, Mason KP, et al. The changing face of homozygous sickle cell disease: 102 patients over 60 years. Int J Lab Hematol. 2009;31(6):585Y596. 4. Platt OS, Brambilla DJ, Rosse WF, et al. Mortality in sickle cell disease. Life expectancy and risk factors for early death. N Engl J Med. 1994;330(23):1639Y1644.
CKD in Patients With Sickle Cell Disease
5. Sklar AH, Campbell H, Caruana RJ, et al. A population study of renal function in sickle cell anemia. Int J Artif Organs. 1990;13(4):231Y236. 6. Abbott KC, Hypolite IO, Agodoa LY. Sickle cell nephropathy at end-stage renal disease in the United States: patient characteristics and survival. Clin Nephrol. 2002;58(1):9Y15. 7. Silva Junior GB, Libo´rio AB, Vieira AP, et al. Evaluation of renal function in sickle cell disease patients in Brazil. Braz J Med Biol Res. 2012;45(7):652Y655. 8. Guasch A, Navarrete J, Nass K, et al. Glomerular involvement in adults with sickle cell hemoglobinopathies: prevalence and clinical correlates of progressive renal failure. J Am Soc Nephrol. 2006;17(8):2228Y2235. 9. Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int Suppl. 2013;3:1Y150. 10. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604Y612. 11. Haymann JP, Stankovic K, Levy P, et al. Glomerular hyperfiltration in adult sickle cell anemia: a frequent hemolysis associated feature. Clin J Am Soc Nephrol. 2010;5(5):756Y761. 12. Falk RJ, Scheinman J, 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(14):910Y915. 13. 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(8):614Y620. 14. Nath KA, Katusic ZS. Vasculature and kidney complications in sickle cell disease. J Am Soc Nephrol. 2012;23(5):781Y784. 15. 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(4):339Y342. 16. Gordeuk VR, Sachdev V, Taylor JG, et al. Relative systemic hypertension in patients with sickle cell disease is associated with risk of pulmonary hypertension and renal insufficiency. Am J Hematol. 2008;83(1):15Y18.
* 2014 The American Federation for Medical Research
Copyright © 2014 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
807
Prevalence and Progression of Chronic Kidney Disease in Adult Patients With Sickle Cell Disease Elvira O. Gosmanova, MD, FASN,* Sahar Zaidi, MD,Þ Jim Y. Wan, PhD,þ and Patricia E. Adams-Graves, MDÞ
Aim: We evaluated the prevalence and progression of chronic kidney disease (CKD) during the 5-year period in a cohort of patients with sickle cell disease (SCD) aged 18 years and older. Methods: We studied 98 patients with SCD. Chronic kidney disease stages I through V were defined based on estimated glomerular filtration rate (eGFR), and albuminuria grades were defined based on spot urine protein-to-creatinine ratio according to the 2012 Kidney Disease Improving Global Outcomes recommendations. In patients with eGFR of greater than 60 mL/min per 1.73 m2, CKD was diagnosed if grade A2 or A3 albuminuria was present. Chronic kidney disease progression was defined as an increase in CKD stage with an additional eGFR reduction of more than 25% from baseline. Results: At baseline, 28.6% of patients had CKD. After a mean follow-up of 5.0 (SD, 0.9) years, 17 patients developed new CKD and the overall CKD prevalence increased to 41.8%. In addition, 8 patients experienced CKD progression. The following baseline variables were associated with the development and progression of CKD in univariate analysis: older age (P = 0.003), higher systolic blood pressure (BP; P = 0.003), lower eGFR (P = 0.001), higher serum creatinine (P = 0.001), and A3 albuminuria (P = 0.008). In multivariate analysis, baseline A3 albuminuria (adjusted odds ratio, 5.0; 95% confidence interval, 1.1-24.3; P = 0.048) and each 1-mm Hg increase in systolic BP (adjusted odds ratio, 1.04; 95% confidence interval, 1.0-1.07; P = 0.039) predicted CKD development and progression. Conclusions: Chronic kidney disease is common in patients with SCD and its prevalence increases with age. Several baseline modifiable and nonmodifiable factors were associated with the development and progression of CKD in patients with SCD. Strategies targeting BP control and proteinuria may be beneficial for individuals with SCD. Key Words: chronic kidney disease, sickle cell disease (J Investig Med 2014;62: 804Y807)
S
ickle cell disease (SCD) is a group of genetic disorders in which a A-globin chain of hemoglobin is affected by several mutations (eg, hemoglobin S, C, A-thalassemia).1 End-organ damage is the major cause of morbidity and mortality in adult patients with SCD.2,3 The survival of patients with SCD is significantly reduced when chronic kidney disease (CKD) is also
From the Divisions of *Nephrology, and †Hematology, Department of Medicine, and ‡Department of Preventive Medicine, University of Tennessee Health Science Center, Memphis, TN. Received November 18, 2013, and in revised form March 3, 2014. Accepted for publication March 14, 2014. Reprints: Elvira O. Gosmanova, MD, FASN, 956 Court Ave, Suite B226, Memphis, TN 38103. E-mail: [email protected]. Disclaimers: None of the authors who participated in the preparation of the article have any potential conflicts of interest. Source of funding: None. This work was not supported by any grant or funding. Copyright * 2014 by The American Federation for Medical Research ISSN: 1081-5589 DOI: 10.1097/01.JIM.0000446836.75352.72
804
present.4,5 Although sickle cell nephropathy is a rare cause of end-stage kidney disease, the survival of these patients is also significantly lower than that of patients with end-stage kidney disease from other etiologies.6 The prevalence of CKD in patients with SCD is dependent on the definitions used to describe the kidney dysfunction. Around 50% of patients with SCD are reported to have estimated glomerular filtration rates (eGFRs) in the hyperfiltration range,7 and 21% to 27% of patients with SCD have eGFR of less than 90 mL/min per 1.73 m2.8 In addition, up to 68% of patients with hemoglobin SS disease and 42% of patients with other sickling hemoglobinopathies were found to have abnormal urinary albumin excretion (in the microalbuminuric or macroalbuminuric range).8 However, to our knowledge, no previous studies have assessed the prevalence of CKD based on the recent Kidney Disease Improving Global Outcomes (KDIGO) recommendations that incorporate eGFR and levels of albuminuria.9 The aim of this retrospective analysis was to evaluate the prevalence and progression of CKD during the 5-year period in a cohort of patients with SCD aged 18 years and older.
METHODS Study Design and Patient Population This was a single-center retrospective cohort study conducted by means of a review of medical records at the Diggs-Kraus Comprehensive Sickle Cell Center, which is affiliated with the University of Tennessee Health Science Center and located at the Regional Medical Center in Memphis, Tenn. We identified patients with SCD who were evaluated at the clinic between January and December 2006 and for whom demographic and clinical data of interest were available from both baseline and follow-up examinations. Individuals with the sickle cell trait were not included. Baseline data collected from the medical records included the patient’s age, sex, race, variant of sickle cell hemoglobinopathy (sickle SS disease and other hemoglobinopathies including sickle SC disease and sickle S-A-thalassemia), serum creatinine (SCr) concentration, serum hemoglobin and total bilirubin levels, urinary protein excretion (UPE), systolic and diastolic blood pressure (BP), hydroxyurea (HU) use, and status with regard to chronic hepatitis B and C as well as human immunodeficiency virus. The follow-up data collected closest to December 2011 included data on SCr concentration and UPE. The final cohort for analysis included 98 patients with SCD who were selected based on the availability of all of the necessary information from both baseline and follow-up. Urinary protein was measured using a colorimetric pyrocatechol violet dye-binding assay (Vitros; Ortho-Clinical Diagnostics, Rochester, NY) and urine creatinine using an enzymatic technique (Vitros; Ortho-Clinical Diagnostics). Urinary protein excretion, expressed as the urine protein-tocreatinine ratio (UPCR) in milligram per gram creatinine, was defined according to the 2012 KDIGO clinical practice guidelines9 as grade A1 or normal albuminuria (UPCR, G150 mg/g), A2 or microalbuminuria (UPCR, between 150-500 mg/g), or A3 Journal of Investigative Medicine
&
Volume 62, Number 5, June 2014
Copyright © 2014 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
Journal of Investigative Medicine
&
Volume 62, Number 5, June 2014
FIGURE 1. Prevalence of CKD stages in patients with SCD.
or macroalbuminuria (UPCR, 9500 mg/g). Kidney function was assessed by estimating glomerular filtration rate (eGFR) as calculated according to the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.10 We evaluated the prevalence of each CKD stage at baseline and after 5 years of follow-up. Chronic kidney disease stages were defined as follows: stage I, eGFR of greater than or equal to 90 mL/min per square meter with the presence of abnormal albuminuria (grade A2 or A3); stage II, eGFR of 60 to 89 mL/min per 1.73 square meter with the presence of abnormal albuminuria; stage III, eGFR of 30 to 59 mL/min per 1.73 square meter; stage IV, eGFR of 15 to 29 mL/min per square meter; and stage V, eGFR of less than 15 mL/min per 1.73 square meter. Hyperfiltration was defined as eGFR of greater than 130 mL/min per 1.73 m2 for women and greater than 140 mL/min per 1.73 m2 for men.11 Chronic kidney disease progression was defined as an increase in CKD stage with an additional 25% reduction in eGFR from baseline.9
Statistical Analysis Statistical analysis was performed using SAS 9.2 (SAS Institute Inc, Cary, NC). All associations with P of less than 0.05 were considered significant. Baseline characteristics were compared using the Mann-Whitney U test and the Fisher exact probability test for continuous and categorical data, respectively. The association between baseline variables and combined outcome with regard to the development and progression of CKD was tested using multivariate regression analysis to model the odds ratios (ORs) with 95% confidence interval (CI) of the cumulative probability of new and worsening CKD among patients with SCD with different baseline characteristics. The following baseline covariates were investigated: age, sex, hemoglobin SS disease versus other hemoglobinopathies, systolic and diastolic BP, baseline SCr, eGFR, baseline serum hemoglobin and total bilirubin concentration, and albuminuria grade. Hepatitis B, hepatitis C, and human immunodeficiency virus were present in 3 patients, 5 patients, and 1 patient, respectively; these patients were not included in multivariate analysis.
RESULTS Among the 98 patients with SCD included in the final analysis, the mean (SD) age for the whole cohort was 31.6 (10.6) years; 43% were men, and 56% had hemoglobin SS disease, whereas 44% had other hemoglobinopathies. At baseline, the mean (SD) SCr was 0.79 (0.42) mg/dL (66.9 [36.9] Kmol/L), the mean (SD) eGFR was 126.0 (32.7) mL/min per 1.73 m2, and 17.3% and 9.2% of patients had grade A2 and A3 albuminuria, respectively. The mean (SD) BP was 120.2/74.0 (15.8/11.8) mm Hg, the mean (SD) hemoglobin was 9.5 (1.9) g/dL, the mean (SD) serum total bilirubin was 3.1 (2.5) mg/dL, and 22% of patients were using HU.
CKD in Patients With Sickle Cell Disease
At baseline, 28 (28.6%) patients met the criteria for CKD diagnosis. The distribution of the various CKD stages is shown in Figure 1. Hyperfiltration was present in 47% of patients. The presence of CKD at baseline was associated with higher SCr (P = 0.001), lower eGFR (P = 0.002), presence of baseline grade A2 or A3 albuminuria (P G 0.001), and lower baseline hemoglobin (P = 0.042; Table 1). The type of hemoglobinopathy, age, sex, baseline total bilirubin concentration, levels of systolic and diastolic BP, and the use of HU were not associated with the presence of CKD (Table 1). After a mean (SD) follow-up of 5 (0.9) years, 17 patients had developed new CKD (defined in all but 1 patient as development of A2 or A3 albuminuria), and the overall prevalence of CKD had increased to 41.8%. The prevalence of stage I CKD had increased from 19.4% to 28.6%, that of stage II CKD had increased from 4.1% to 8.2%, and that of stage IV CKD had increased from 1% to 2%, whereas that of stage III CKD had decreased from 4.1% to 1%. During this time, 2% of patients progressed to CKD stage V (Fig. 1). At the end of the follow-up period, 65.4% of patients with abnormal albuminuria at baseline had retained stable UPE, whereas 11.5% of patients had progressed from grade A2 albuminuria to grade A3, and 23.1% of patients had regressed from a higher grade of albuminuria to a lower grade (2/3 of these from A2 to A1 and 1/3 of these from A3 to A2). Among the patients with normal albuminuria at baseline, 17 (17.3%) had developed abnormal albumin excretion (10 patients developed grade A2 albuminuria, and 7 patients developed grade A3). In unadjusted analysis, older age (P = 0.003), higher baseline systolic BP (P = 0.003), higher baseline SCr (P = 0.001), lower eGFR (P = 0.001), and A3 albuminuria (P = 0.008) were associated with combined outcome of CKD development or progression (Table 2). In multivariate analysis that excluded albuminuria, each decrease of 10 mL/min per 1.73 m2 in eGFR was associated with increased risk of CKD development or progression (adjusted OR, 1.49; 95% CI, 1.10-2.01; P = 0.01). In adjusted analysis that excluded eGFR, the presence of A3
TABLE 1. Baseline Characteristics Associated With the Presence of CKD in Patients With SCD CKD Patients Baseline Patient Characteristics Age, mean (SD), y Sex, male, % Race, African American, % Hemoglobin SS disease, % SCr, mean (SD), mg/dL eGFR, mean (SD), mL/min per 1.73 m2 UPCR, 90.15 mg/g, % Serum hemoglobin, mean (SD), g/dL Total bilirubin, mean (SD), mg/dL SBP, mean (SD), mm Hg DBP, mean (SD), mm Hg HU use, %
Yes (n = 28)
No (n = 70)
P
33.8 (11.9) 57.1 96.4 60.7 1.0 (0.66) 112.3 (43.2)
30.5 (9.6) 38.6 97.1 52.9 0.7 (0.2) 133.8 (23.1)
NS NS NS NS 0.001 0.002
92.8 9.1 (1.9)
0 9.8 (1.7)
G0.001 0.042
3.7 (2.7)
2.9 (2.4)
NS
124.5 (12.6) 76.8 (12.6) 32.1
118.6 (16.2) 73.0 (11.5) 21.4
NS NS NS
DBP indicates diastolic BP; NS, not significant; SBP, systolic BP.
* 2014 The American Federation for Medical Research
Copyright © 2014 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
805
Journal of Investigative Medicine
Gosmanova et al
TABLE 2. Risk Factors Associated With CKD Development or Progression in Patients With SCD CKD Development or Progression Patients Baseline Patient Characteristics
Yes (n = 25)
No (n = 73)
Age, mean (SD), y 36.7 (11.1) 29.7 (9.5) Sex, male, % 52.0 41.2 Hemoglobin SS disease, % 56.0 51.2 SCr, mean (SD), mg/dL 1.0 (0.7) 0.7 (0.2) 109.5 (42.5) 133.9 (24.2) eGFR, mean (SD), mL/min per 1.73 m2 UPCR, 90.5 mg/g, % 24 4.1 Serum hemoglobin, 9.6 (2.1) 9.6 (1.6) mean (SD), g/dL Total bilirubin, mean (SD), g/dL 3.3 (3.0) 3.0 (2.3) SBP, mean (SD), mm Hg 128.4 (19.9) 117.6 (11.4) DBP, mean (SD), mm Hg 78.0 (13.0) 72.8 (11.3) HU use, % 28.0 21.9
P 0.003 0.3 0.9 0.001 0.001 0.008 0.9 0.7 0.003 0.062 0.5
DBP indicates diastolic BP; SBP, systolic BP.
albuminuria (OR, 5.0; 95% CI, 1.1Y24.3; P = 0.048) was predictive of CKD development and progression. In addition, each 1-mm Hg increase in systolic BP was associated with an OR of 1.04 (95% CI, 1.01Y1.07; P = 0.039) for combined outcome of the development or progression of CKD. The associations between systolic BP and abnormal albuminuria and CKD development or progression were attenuated by adjustment for eGFR (P 9 0.1), suggesting that eGFR may be an intermediate variable among these risk factors and the combined renal outcome of CKD development and progression.
DISCUSSION The concomitant presence of CKD adversely affects the health outcomes of patients with SCD. Patients with SCD and elevated SCr have reduced survival compared with patients with SCD with no nephropathy.4,5 Therefore, it is important to understand the prevalence and progression of CKD in these patients. To our knowledge, this is the first study to evaluate the prevalence of various CKD stages in patients with SCD based on the KDIGO guidelines combining eGFR and levels of albuminuria.9 It has been reported that 4% to 7% of patients with SCD have elevated SCr.12,13 As in a study from Brazil, we found that 5.1% of patients with SCD had an eGFR of less than 60 mL/min per 1.73 m2 using the CKD-EPI equation.7 Although the overall number of patients with moderate to severe CKD was small, 80% of them (4/5 patients) progressed to a more advanced CKD stage during the 5-year period. Another study used eGFR of less than 90 mL/min per 1.73 m2, a value derived from the CockroftGault equation, as the threshold for CKD, and it found CKD to be present in 21% of patients with SCD, an incidence that was higher than the 13% seen in the current study, applying a similar eGFR threshold.8 As recommended by the KDIGO, we used the CKD-EPI equation as the preferred method of estimating GFR in the absence of data, suggesting that other equations would be superior for estimating GFR in patients with SCD.9 The CKDEPI equation has been shown to be more accurate in the estimation of true GFR with values above 60 mL/min per 1.73 m2.10
806
&
Volume 62, Number 5, June 2014
Similar to previous reports,7,11 we found that hyperfiltration was present in 47% of patients with SCD. We observed that lower hemoglobin concentration was associated with the presence of hyperfiltration (data not shown). A link was recently proposed between hemolysis that results in lower serum hemoglobin concentrations in patients with SCD and hyperfiltration.14 The heme released during hemolysis of sickling red blood cells can activate an inducible hemoxygenase and increase the formation of carbon monoxide, which, in turn, causes renal vasodilatation, increased intraglomerular pressure, and subsequent hyperfiltration.14 A total of 20.4% of patients with SCD with eGFR in the hyperfiltration range had increased albuminuria (A2 or A3). These data suggest that a large number of patients with normal and elevated GFR meet the KDIGO definition for the presence of stage I CKD. Several relatively small and short-term studies have evaluated the antiproteinuric effect of angiotensin converting enzyme inhibitors in patients with SCD12,15 and have shown that angiotensin converting enzyme inhibitors effectively reduce urinary albumin excretion; nevertheless, long-term studies are needed to demonstrate whether this reduction in albuminuria can also result in a slowing of CKD progression in patients with SCD. Previously, a high incidence (68%) of abnormal albuminuria, determined from the urine albumin-to-creatinine ratio, was reported in patients with SCD.8 In the present study, the prevalence of abnormal albuminuria at baseline was lower at 26.5% but subsequently increased to 42.8% after 5 years of follow-up. There is an important methodological difference between the albuminuria measurement techniques used in the study by Guasch et al.8 and the current study. The former study used the albumin-to-creatinine ratio for albuminuria classification. We used the UPCR to determine the albuminuria grades (A1-normal, A2, or microalbuminuria, and A3, or macroalbuminuria) based on the 2012 KDIGO recommendations, which support UPCR as an acceptable alternative method for albuminuria gradation when direct urine albumin measurements are not available.9 We were also able to analyze the natural history of albuminuria progression. We observed that during the 5-year follow-up period, most of the patients with abnormal albuminuria at baseline (65.4%) had unchanged UPE, whereas a smaller number (11.5%) of patients progressed to a higher albuminuria grade, and 23.1% of patients regressed to a lower albuminuria grade (most [2/3] from A2 to normal but some [1/3] from A3 to A2 albuminuria). After 5 years of follow-up, 17.3% of patients developed new albuminuria. It has previously been shown that older age and higher systolic BP are associated with the presence of albuminuria.8 We also found that older age and higher baseline systolic BP were associated with the separate outcome of new albuminuria development (data not shown). Age is not a modifiable factor, but careful monitoring and treatment of elevated BP to reduce the likelihood of developing CKD are potentially achievable goals. The difficulty lays in determining optimal BP values, above which BP should be treated, for patients with SCD. The average systolic and diastolic BP in patients with SCD are similar to those in the general population in individuals aged 18 to 45 years; in individuals aged older than 45, however, systolic and diastolic BP tend to be lower in patients with SCD than in similar individuals without SCD.8 Forty-four percent and ten percent of patients with SCD are reported to have systolic and diastolic BP between 120 to 140/80 to 90 mm Hg and greater than 140/90 mm Hg, respectively.16 Despite the fact that most of the patients have BP values below the hypertensive range, BP values in the prehypertension range (120Y140/ 80Y90 mm Hg) have been shown to be adversely associated with * 2014 The American Federation for Medical Research
Copyright © 2014 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
Journal of Investigative Medicine
&
Volume 62, Number 5, June 2014
the presence of pulmonary hypertension and elevated SCr16; this suggests that BP levels greater than 120/80 mm Hg might be considered ‘‘relative’’ hypertension in patients with SCD. We also found that even small increases in systolic BP were associated with increased risk of CKD development and progression. Before routine antihypertensive treatment can be instituted for BP of greater than 120/80 mm Hg, however, this hypothesis requires further testing in prospective trials. Several important limitations of the present study need to be recognized. This was a single-site observation study with a relatively small study population. Not all potential confounders that may have influenced the development of renal dysfunction in patients with SCD were investigated; one notable confounder that was left out is >-thalassemia status, which may be protective against kidney dysfunction. We also used UPCR to characterize albuminuria, whereas urine albumin to creatinine ratio is more sensitive in the detection of low levels of urinary albumin (G300 mg/d). The main strength of this study is that it is the first to provide information about the prevalence of the various CKD stages and their progression in patients with SCD. In conclusion, CKD in patients with SCD is common and its prevalence increases with age. Chronic kidney disease was observed in 35.3% of patients with SCD aged younger than 40 years old and in 53.3% of patients with SCD aged older than 40. Strategies targeting control of BP and proteinuria may be beneficial for individuals with SCD. REFERENCES 1. Kato GJ, Hebbel RP, Steinberg MH, et al. Vasculopathy in sickle cell disease: Biology, pathophysiology, genetics, translational medicine, and new research directions. Am J Hematol. 2009;84(9):618Y625. 2. Powars DR, Chan LS, Hiti A, et al. Outcome of sickle cell anemia: a 4-decade observational study of 1056 patients. Medicine (Baltimore). 2005;84(6):363Y376. 3. Serjeant GR, Serjeant BE, Mason KP, et al. The changing face of homozygous sickle cell disease: 102 patients over 60 years. Int J Lab Hematol. 2009;31(6):585Y596. 4. Platt OS, Brambilla DJ, Rosse WF, et al. Mortality in sickle cell disease. Life expectancy and risk factors for early death. N Engl J Med. 1994;330(23):1639Y1644.
CKD in Patients With Sickle Cell Disease
5. Sklar AH, Campbell H, Caruana RJ, et al. A population study of renal function in sickle cell anemia. Int J Artif Organs. 1990;13(4):231Y236. 6. Abbott KC, Hypolite IO, Agodoa LY. Sickle cell nephropathy at end-stage renal disease in the United States: patient characteristics and survival. Clin Nephrol. 2002;58(1):9Y15. 7. Silva Junior GB, Libo´rio AB, Vieira AP, et al. Evaluation of renal function in sickle cell disease patients in Brazil. Braz J Med Biol Res. 2012;45(7):652Y655. 8. Guasch A, Navarrete J, Nass K, et al. Glomerular involvement in adults with sickle cell hemoglobinopathies: prevalence and clinical correlates of progressive renal failure. J Am Soc Nephrol. 2006;17(8):2228Y2235. 9. Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int Suppl. 2013;3:1Y150. 10. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604Y612. 11. Haymann JP, Stankovic K, Levy P, et al. Glomerular hyperfiltration in adult sickle cell anemia: a frequent hemolysis associated feature. Clin J Am Soc Nephrol. 2010;5(5):756Y761. 12. Falk RJ, Scheinman J, 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(14):910Y915. 13. 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(8):614Y620. 14. Nath KA, Katusic ZS. Vasculature and kidney complications in sickle cell disease. J Am Soc Nephrol. 2012;23(5):781Y784. 15. 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(4):339Y342. 16. Gordeuk VR, Sachdev V, Taylor JG, et al. Relative systemic hypertension in patients with sickle cell disease is associated with risk of pulmonary hypertension and renal insufficiency. Am J Hematol. 2008;83(1):15Y18.
* 2014 The American Federation for Medical Research
Copyright © 2014 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
807
