Patients with sickle cell disease who develop end‐stage kidney disease continue to experience poor survival — A 19‐year United States Renal Data System study

Patients with sickle cell disease (SCD) experience increased morbidity and early mortality, due in part to organ damage, including end-stage kidney disease (ESKD).1 Decline in kidney function is progressive with initial hyperfiltration, followed by declines in glomerular filtration rate (GFR) and eventual ESKD in adults. Individuals with SCD experience a more rapid decline in GFR than the general African American population and this rapid decline is associated with increased mortality.2, 3 Prior studies suggest the one-year mortality rate after initiating dialysis is higher in patients with SCD but early care by a nephrologist may decrease mortality.4 Additionally, individuals with ESKD due to SCD (SCD-ESKD) are less likely to receive an arteriovenous fistula (AVF) when initiating haemodialysis (HD)5, 6 despite studies demonstrating improved survival when dialyzed with an AVF or graft.7, 8 Finally, referral for transplantation and transplantation rates are lower among individuals with SCD-ESKD despite the improved survival benefit of kidney transplantation.9-11 It is important to characterize the mortality of and disparities experienced by patients with SCD-ESKD. These data can highlight the need to improve clinical care, monitor kidney disease progression, and develop therapies to prevent SCD-ESKD. Despite improvement in care over the last two decades, the prognosis of SCD-ESKD remains uncertain. This study aimed to characterize the outcomes of individuals with SCD-ESKD from 1998 toi 2017 enrolled in the United States Renal Data System (USRDS) registry. We hypothesized that survival of patients with SCD-ESKD has improved over the course of the last two decades. We performed an institutional review board (IRB)-approved retrospective analysis using the USRDS database from 1998 to 2017. Data including age, sex, race, ethnicity, primary diagnosis, outcome dates, first ESKD treatment modality, and comorbid conditions (atherosclerotic heart disease, congestive heart failure, cerebrovascular disease, diabetes mellitus, peripheral vascular disease and hypertension) were abstracted from the patient profile using Centers for Medicare and Medicaid Services (CMS) Medical Evidence Form 2278. Age was summarized by mean (standard deviation), median (interquartile range), and by number and percentage within age groups. Sex, ethnicity, first ESKD modality, dialysis access, duration of pre-ESKD nephrology care, and comorbid conditions were summarized by number and percentage (Table 1). Groups were divided first by SCD status and then by Black race. Adjusted (Figure 1) and unadjusted (Figure S1) survival and time-to-transplant were calculated using the Cox proportional hazards regression model, overall and by year of first ESKD event. For survival, adjusted values were adjusted for age, sex, ethnicity, Black race, initial treatment modality, haemodialysis access type, pre-ESKD nephrology care and comorbidities. For transplant, due to small numbers in the temporal analysis, adjustment was restricted to demographic variables including age (on a continuous scale), sex, ethnicity and Black race. Between 1998 and 2017, 2 194 079 patients developed ESKD; 2018 (0.09%) had a diagnosis of SCD. The age at ESKD diagnosis was significantly lower in the SCD-ESKD group than in the non-SCD-ESKD group, both unadjusted [overall response (OR), 0.57; 95% confidence interval (CI), 0.55–0.58, for each 10-year increase in age] and when adjusted for sex, race, ethnicity, initial treatment modality, access, nephrology care and comorbidities (OR, 0.62; 95% CI, 0.60–0.64, for each 10-year increase in age). Twenty percent of patients developed ESKD before the age of 30. Male patients with SCD-ESKD developed ESKD at an earlier age than female patients with SCD-ESKD (p < 0.001). Patients with SCD-ESKD had lower adjusted odds of receiving a pre-emptive kidney transplant as compared to being started on dialysis (OR, 0.24; 95% CI, 0.14, 0.41). Patients with SCD-ESKD were more likely to have ongoing pre-ESKD nephrology care than not having any care (<6 months: OR, 1.23; 95% CI, 1.01–1.48; 6–12 months: OR, 1.33; 95% CI, 1.13–1.58; and > 12 months: OR, 1.43; 95% CI, 1.22–1.69). The unadjusted hazard ratio of mortality following first ESKD event was higher among SCD-ESKD patients than non-SCD-ESKD patients [hazard ratio (HR), 1.21; 95% CI, 1.15–1.27]. This hazard of mortality effect was much larger after adjusting for age, sex, ethnicity, Black race, initial treatment modality and comorbidities (HR: 2.68, 95% CI: 2.55–2.83). Over the 19-year study period, the adjusted hazard of mortality improved by 35% (HR 0.65, 95% CI 0.64–0.66) for non-SCD-ESKD and by 30% (HR 0.70, 95% CI 0.58–0.84) for SCD-ESKD. These values were not significantly different between the groups. The unadjusted rate of receiving a transplant was lower among SCD-ESKD patients than non-SCD-ESKD patients by 29% (OR, 0.71; 95% CI, 0.62–0.83). This relative transplant rate was even smaller after adjusting for age, sex, ethnicity, and Black race (HR, 0.49; 95% CI, 0.29–0.84). Over the study period, the adjusted rate of transplant decreased for the non-SCD-ESKD (HR, 0.81; 95% CI, 0.80–0.82), but this decrease was not statistically significant for the SCD-ESKD group (HR, 0.75; 95% CI, 0.43–1.32). Patients with SCD continue to develop ESKD earlier in life, have a higher risk of death, and lower transplant rate than non-SCD-ESKD patients. When adjusting for age, this disparity in survival was magnified as SCD-ESKD patients have a 2.7 times higher hazard of mortality. In contrast to a prior study,4 we found that despite the increase in pre-ESKD nephrology care, mortality remained elevated. It remains alarming that patients with SCD-ESKD were less likely to receive a kidney transplant as compared to non-SCD-ESKD patients. Data remain conflicting as to whether SCD patients have lower overall and graft survival than non-SCD patients; this should not preclude patients from transplant.11-13 This retrospective study has some limitations. We could not assess the stage of chronic kidney disease at time of referral. USRDS reports active medications, but we could not evaluate the effect of SCD-disease modifying or renoprotective therapies on progression to ESKD. Large prospective cohort studies are needed to determine the impact of therapies on progression of kidney disease. This study confirms that patients with SCD who develop ESKD continue to experience disparities in renal outcomes. As patients with SCD-ESKD are at high risk of all-cause mortality, it is important to ensure comprehensive SCD care centres that include both paediatric and adult haematologists.14 Ongoing research should identify risk factors for ESKD progression, ensure implementation of monitoring for kidney disease, and determine therapies that prevent the development of ESKD. Project conceptualization: Jeffrey C. Winer, Rima S. Zahr; data acquisition: Jeffrey C. Winer, Rima S. Zahr; analysis: Jeffrey C. Winer; interpretation: Jeffrey C. Winer, Rima S. Zahr, Kenneth I. Ataga, KL, Marianne E. Yee; project administration: Rima S. Zahr; writing original draft: Jeffrey C. Winer, Rima S. Zahr; writing review and editing: Jeffrey C. Winer, Rima S. Zahr, Kenneth I. Ataga, Jeffrey D. Lebensburger. We thank the USRDS for approving our study and giving us the opportunity to carry out this study. Rima S. Zahr receives research funding from the NIH/NHLBI K23HL157554. Rima S. Zahr, Jeffrey C. Winer, Marianne E. Yee: no conflict of interests; Kenneth I. Ataga: no relevant conflict of interests related to this manuscript, but has served on advisory boards for Novartis, Global Blood Therapeutics, Novo Nordisk, Roche and Forma Therapeutics; Jeffrey D. Lebensburger: consultant for Novartis, Agios and Forma Therapeutics for studies unrelated to this manuscript. This study was approved by the institutional review board at the University of Tennessee Health Science Center at Memphis and the United States Renal Data System (USRDS). Data are made available by contacting the USRDS. Figure S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.