Predictors of nephropathy in sickle cell disease: A cross-sectional analysis across age and clinical phenotypes in Zaria, Nigeria

Abstract

Background:

Sickle cell disease (SCD) is associated with significant complications, including nephropathy, which can progress to end-stage renal disease. While risks for SCD nephropathy have been identified globally, their specific prevalence and predictors in Nigerian, especially across age and clinical phenotypes, are less well understood. This study assessed the prevalence and predictors of nephropathy among Nigerian SCD patients.

Methods:

This was a cross-sectional study among paediatrics, adolescent, and adult patients with SCD attending hematology clinics of Ahmadu Bello University Teaching Hospital (ABUTH), Zaria. Interviewer-administered questionnaires were utilised to collect demographic and clinical data. Nephropathy was defined by the presence of proteinuria, haematuria, or eGFR <60 ml/min. Patients were classified into hyper-haemolytic and vaso-occlusive sub-phenotypes. Statistical analyses involved Mann-Whitney U tests, chi-square tests, and logistic regression, with a significance level set at p < 0.05.

Results:

A total of 311 SCD patients were enrolled, with a median age of 17 (IQR 9–25) years and 65% female. Nephropathy was present in 13.8% of patients, with significantly higher prevalence among adults (23.7%) compared to adolescents (9.1%) and paediatrics patients (3.8%) respectively (X²= 19.27, p < 0.001). The odds of nephropathy were significantly higher in adults (OR 5.57, 95% CI: -0.53–3.96) and females (OR 2.05, 95% CI: 1.72–3.16). Patients with hyper-haemolytic complications had higher odds of nephropathy than those with a vaso-occlusive phenotype (P = 0.366).

Conclusion:

The prevalence of nephropathy in Nigerian SCD patients increases with age and is more common in females, with hyper-hemolytic sub-phenotypes showing a higher risk profile.

Introduction

Sickle cell disease (SCD) is an inherited hemoglobinopathy associated with complications arising from chronic haemolysis, vaso-occlusion, and systemic multi-organ involvement. Among these complications, nephropathy is a severe and common manifestation that can significantly impact morbidity and mortality in SCD patients [1]. Studies globally have shown that SCD-related nephropathy prevalence and risk factors vary across populations. Research from West Africa, where SCD prevalence is high, has highlighted an increased burden of nephropathy in SCD, often with earlier onset compared to Western populations [2]. Studies in Nigeria and other West African countries have demonstrated that age, gender, and clinical phenotypes such as haemolytic and vaso-occlusive subtypes influence nephropathy risk [3,4].

In high-income countries such as the United States and Europe, the availability of comprehensive care has improved SCD management, yet nephropathy remains a significant concern. Studies from the U.S. have shown that the prevalence of proteinuria and reduced eGFR in adult SCD patients ranges from 20% to 30%, often exacerbated by delayed diagnosis and suboptimal control of haemolysis [1]. Similar reports from Europe show the same trends in nephropathy risk, although earlier intervention appears to delay progression to end-stage renal disease (ESRD) [5,6]. In Asia, where SCD is less prevalent, studies have documented lower nephropathy rates, likely due to variations in genetic expression and the environment [7].

This study aimed to explore the prevalence and predictors of nephropathy in SCD patients across paediatrics, adolescent, and adult age groups in Zaria, Nigeria. By stratifying patients by age and examining clinical phenotypes, we sought to better understand how demographic and clinical factors contribute to nephropathy risk in an African context, providing data for comparison with other global studies. This could also guide the introduction of targeted screening protocols for nephropathy across demographic strata of people with SCD in Nigeria.

Methodology

Following the acquisition of Institutional ethical (Health Research Ethics Committee of ABUTH Zaria-Reference No ABUTHZ/HREC/W48/2022) approval, this cross-sectional study enrolled consenting adults, assented children, and adolescent participants with sickle cell disease. Participants were enrolled at paediatric and adult Haematology sickle cell clinics of ABUTH Zaria. Interviewer-administered questionnaires were utilized, and data were collated into a database. Blood and urine samples were also utilized for serum creatinine and urinalysis. Estimated creatinine clearance was computed using eGFR (CKD-EPI for adults and Schwartz for children and adolescents).

Participants were stratified by age as paediatric (<10 years), adolescents (10 – 19 years), and adults (>19 years). Nephropathy was defined as any measure of proteinuria ± haematuria and ± eGFR <60 ml/min. Participants with complications due to excess chronic hyper-haemolysis, such as leg ulcers, priapism, stroke [8], and the presence of gallstones, were grouped as having a high likelihood of a haemolytic sub-phenotype, while osteonecrosis [8] was categorized as a vaso-occlusive pain sub-phenotype.

Data Analyses

Data was analyzed using both Jeffrey’s Amazing Statistics Program (JASP) version 0.16.4 (University of Amsterdam) and Windows Programme for Epidemiologists (WINPEPI version 11.65). Continuous variables were summarized as medians and interquartile ranges (IQRs) after assessing distributions using skewness, kurtosis, and visual inspections of histograms and Q-Q plots. Differences in the distributions of these continuous variables between qualitative variables with two categories were assessed using Mann-Whitney U (MWU) tests. Chi-Square tests were used to assess the relationship between age categories and nephropathy, and subsequent pairwise comparisons were done. Logistic regression analysis was conducted to predict the presence of nephropathy using predictors that have statistically significant relationships at the level of bivariate analyses. Odds ratios with 95% Confidence Intervals (CI) were computed. Two-tailed hypotheses and a critical alpha level of 0.05 were set for all analyses.

Results

A total of 313 participants were enrolled in this study; however, only participants with a complete set of information were included for each statistical analysis. The median (IQR) age of the study participants was 17.0 (9.0, 25.0) years. Females constituted 202/311 (65.0%) of the participants. Participants with HbSS were 289/311(92.9%) while HbSS+F and HbSC were 7(2.3%) and 15(4.8%) respectively. Paediatric patients constituted 80/311(25.7%), adolescents 10 to 19 years were 99/311(31.8%) and adults 132/311(42.4%). Nephropathy was present in 43/311 (13.8%) of the participants. There was a statistically significant association between age (paediatric, adolescent, and adult) and nephropathy (present vs absent); with nephropathy present in 3/80 (3.8%), 9/99(9.1%), and 31/131 (23.7%) of paediatric, adolescent, and adult participants, respectively (X²=19.269, p<0.001). Pairwise comparisons of age categories are summarized in Table 1; Adolescents had higher odds of having nephropathy compared to paediatric participants. Also, adults had markedly higher odds of having nephropathy compared to either pediatrics or adolescents.

Table 1:

Comparison of risk of Nephropathy by age among SCD patients

Age Categories.	Odds ratio (95% CI)	Chi-sq.	P-Value
Adolescent vs Paediatrics	2.567 (0.671 to 9.818)	2.018	0.155
Adult vs Paediatrics	7.957 (2.345 to 26.998)	14.572	0.0001
Adult vs Adolescents	3.100(1.400 to 6.864)	8.335	0.004

# Sidak-adjusted (Bonferroni in brackets)

A summary of the clinical phenotypes is shown in Figure 1. The median(IQR) age of participants with nephropathy was statistically significantly higher than that of those without nephropathy {24.0(19.0, 32.0)years vs 15.0(8.0, 23.0) years, MWU=8333.5000, p<0.001}.

Figure 1.

Distribution of Clinical Phenotypes

The odds of a female having nephropathy were higher than those of a male (OR 3.849, 95% CI: 1.570, 9.440). Additionally, participants with complications associated with haemolysis had higher but statistically non-significant odds of having nephropathy compared to those with vaso-occlusion.

A binary logistic regression model was constructed with nephropathy (present vs absent) as the outcome variable and gender (female vs male) as well as arm of study (paediatrics, adolescent, adults) as predictor variables. The adjusted ORs were female (2.053, 95% CI; -1.723, 3.161, p=0.564), adults (5.57, 95% CI; - 0.527, 3.962, p=0.134), and adolescents (1.08, 95% CI; -2.729, 2.829, p=0.955)

Table 2:

Clinical events in the previous 12-month

	Pain Crises	Blood Transfusions	Hospitalizations
25th percentile	0.00	0.00	0.00
50th percentile	2.00	0.00	0.00
75th percentile	3.00	1.00	1.00

Table 3:

Relationship between age, sickle cell crisis, hospitalization, and nephropathy

	Age(years) vs Nephropathy		NOP vs Nephropathy		NOB vs Nephropathy		NOH vs Nephropathy
	Absent	Present	Absent	Present	Absent	Present	Absent	Present
25th percentile	8.00	19.00	0.00	0.50	0.00	0.00	0.00	0.00
50th percentile	15.00	24.00	2.00	1.00	0.00	0.00	0.00	0.00
75th percentile	23.00	32.00	3.00	2.00	1.00	1.00	1.00	1.00
MWU		3104.50	6207.00		5254.50		5752.00
p-value		< .001	0.38		0.40		0.84

NOP: No of painful crises, NOB: No of blood transfusions, NOH: No of hospitalizations per year

Table 4:

Relationship between gender, clinical sickle cell phenotypes, and nephropathy

		Nephropathy		X2	p-value	OR (95% CI)
		Present	Absent
Gender	Female	37(18.3 %)	165(81.7 %)			3.849
	Male	6(5.5 %)	103(94.5 %)	9.754	0.002	(1.570, 9.440)
Phenotype	Haemolytic	8(18.2%)	36(81.8%)			1.473
	Vaso-occlusive	35(13.1%)	232(86.9%)	0.816	0.366	(0.633, 3.427)

Discussion

The findings of this study highlight a significant prevalence of nephropathy in Nigerian SCD patients, particularly among adults, aligning with reports from other West African countries. In Southern Nigeria, studies [3,4] have similarly shown that nephropathy prevalence increases with age, with rates among adults often exceeding 20%. This prevalence is higher than that reported in Asian studies, where genetic and environmental differences may influence SCD's clinical progression [7]. However, it remains comparable to rates observed in high-income countries like the U.S. and Europe, where adult nephropathy prevalence ranges between 20% and 30% [1]

Gender differences observed in our study, with higher nephropathy odds among females, are supported by Okpere et al. [4] and contrast with findings from some U.S. studies where gender differences are less pronounced [9]. This variation might indicate an interaction between genetic, hormonal, and environmental factors unique to the West African population. Furthermore, the association between clinical phenotype and nephropathy risk in this study aligns with global findings. Hemolytic subtypes, linked with more severe disease courses, have been associated with higher nephropathy risk in studies from both African and Western populations [8,10].

The odds of nephropathy among adult Nigerian patients in our study were markedly higher compared to younger age groups, suggesting a cumulative effect of chronic hemolysis and microvascular damage over time. This trend mirrors observations in the U.S. and Europe, where adults with prolonged SCD exposure are at higher risk for chronic kidney disease (CKD) progression [6]. However, in Nigeria, early onset of nephropathy may be more common due to limited healthcare access and lower baseline GFRs, exacerbating risks compared to Western cohorts [3].

Conclusion

Overall, this study supports the idea that nephropathy risk in SCD varies by age, gender, and clinical phenotype and underscores the need for regional comparisons. Tailored interventions to prevent nephropathy progression, especially in high-risk groups such as females and adults, are crucial for improving SCD outcomes in Nigeria and West Africa. This study also highlights the need for further research comparing regional trends to better understand the mechanisms driving nephropathy risk in different populations.

Conflict of interest

Nil