Trends in the epidemiology of childhood nephrotic syndrome in Africa: A systematic review

Summary

Background

Childhood nephrotic syndrome, if left untreated, leads to progressive kidney disease or death. We quantified the prevalence of steroid-sensitive nephrotic syndrome, steroid-resistant nephrotic syndrome, and histological types as the epidemiology of nephrotic syndrome in Africa remains unknown, yet impacts outcomes.

Methods

We searched MEDLINE, Embase, African Journals Online, and WHO Global Health Library for articles in any language reporting on childhood nephrotic syndrome in Africa from January 1, 1946 to July 1, 2020. Primary outcomes included steroid response, biopsy defined minimal change disease, and focal segmental glomerulosclerosis (FSGS) by both pooled and individual proportions across regions and overall.

Findings

There were 81 papers from 17 countries included. Majority of 8131 children were steroid-sensitive (64% [95% CI: 63–66%]) and the remaining were steroid-resistant (34% [95% CI: 33–35%]). Of children biopsied, pathological findings were 38% [95% CI: 36–40%] minimal change, 24% [95% CI: 22–25%] FSGS, and 38% [95% CI: 36–40%] secondary causes of nephrotic syndrome.

Interpretation

Few African countries reported on the prevalence of childhood nephrotic syndrome. Steroid-sensitive disease is more common than steroid-resistant disease although prevalence of steroid-resistant nephrotic syndrome is higher than reported globally. Pathology findings suggest minimal change and secondary causes are common. Scarcity of data in Africa prevents appropriate healthcare resource allocation to diagnose and treat this treatable childhood kidney disease to prevent poor health outcomes.

Funding

Funding was provided by the Canadian Institute for Health Research (CIHR) and the National Institute of Health (NIH) for the H3 Africa Kidney Disease Research Network. This research was undertaken, in part, from the Canada Research Chairs program.

Highlights

• •There are scarce data on childhood nephrotic syndrome across Africa with only 31% of African countries reporting.
• •Mean age of diagnosis of childhood nephrotic syndrome was 7.3 years which is higher than reported in Europeans and Asians
• •Most children (64%) have steroid sensitive disease and access to steroids and repeated courses is needed
• •About 34% of children are steroid resistant and need access to second line medications
• •Kidney biopsy should occur more often in African children as the proportion with secondary causes is much higher

Introduction

Childhood nephrotic syndrome is a glomerular disorder with expected good clinical outcomes [1,2]. If left untreated, it may affect kidney function, or in children who have frequent relapses or are non-responsive to treatment, it may progress to chronic kidney disease, end stage kidney disease, or lead to early mortality [1]. While the cause of idiopathic childhood nephrotic syndrome remains unknown, treatment with corticosteroids (steroids) has greatly reduced the morbidity and mortality. Less than 5% of children who are non-responsive to steroid treatment progress to end-stage kidney disease in high-resource countries [1]. Response to steroid treatment, however, is variable depending on ethnicities and geographic regions [3]. Steroid-resistant nephrotic syndrome is associated with a worse prognosis and is a known risk factor for progression to end-stage kidney disease [1]. In the United States, African Americans have three times the prevalence of steroid-resistant nephrotic syndrome compared to European American children [3,4].

Among Africans, steroid resistance in children ranges from 16% to 73·5% [2,3]. In tropical Africa, incidence of nephrotic syndrome is 0·35–1·34% of hospital admissions, though studies from northern and southern African countries are not available [2]. Most studies are cross-sectional with few longitudinal studies, limited to single centers. As the cohort size is typically small, a pooled analysis is needed to understand the epidemiology and trends across the African continent. Recent reports suggest a changing epidemiology with more children responding to steroid treatment and lower rates of focal segmental glomerulosclerosis (FSGS), presumably due to the decline in quartan malaria infections and improved access to healthcare [5]. In tropical Africa, pre-1989, when quartan malaria nephropathy was the most common glomerular lesion reported among children, rates of steroid resistance were 73·5% and after 1989, declined to 27·4%, likely due to improved and affordable malaria treatment [2,5].

Rising rates of noncommunicable diseases in Africa, especially chronic kidney disease among children and young adults, makes understanding the epidemiology of childhood nephrotic syndrome in Africa crucial. If left unrecognized or untreated nephrotic syndrome can lead to chronic kidney disease/end-stage kidney disease which confers significant morbidity and mortality. Moreover, prevention of progression towards end-stage kidney disease is paramount, as renal replacement therapy is prohibitive to many health systems as well as the individuals and families. Updated information on the epidemiology and trends of nephrotic syndrome in children will ensure informed decisions on recognition, diagnosis and appropriate management of potentially treatable forms of childhood kidney disease. This systematic review was conducted to determine the proportions of steroid-sensitive nephrotic syndrome, steroid-resistant nephrotic syndrome, and its histological types among African children to understand how they impact outcomes.

Methods

We followed the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines. Our search strategy involved a MEDLINE search from 1946 to July 1st 2020 and an Embase search from 1947 to July 1st 2020. Search terms included “nephrotic syndrome”, or “edema hypertension proteinuria syndrome”, or “oedema hypertension proteinuria syndrome” as well as “pediatric”, “adolescent”, and every country within the African continent (Supplementary File). African Journals Online and WHO Global Health Library were also searched using key terms “nephrotic syndrome”, and “childhood” to find additional studies. References of any review articles were cross referenced to supplement additional relevant articles.

Eligibility criteria included studies in any language that contained primary data on children with nephrotic syndrome ages 0–18 years in any African country. Studies were excluded if the study population included only adult participants, were from outside of Africa and if the diagnosis was not nephrotic syndrome or only systemic diseases associated with nephrotic syndrome were reported. Single case reports were also excluded. Published studies were screened based on the title and abstract and if they met the inclusion criteria, the full text was reviewed if accessible.

Papers were assessed, screened, and abstracted (RW) with additional review by co-authors (investigators in the H3 Africa Kidney Disease Research Network). Any discrepancies were reviewed and adjudicated by two authors (JVR, RSP). Data collected included the country of study, age at diagnosis, sex, number of overall study participants, number with nephrotic syndrome, proportion of children treated with steroids, number of children with steroid-resistant or steroid-sensitive nephrotic syndrome, number of children biopsied, outcome of biopsies, and study period. For studies with overlapping patient populations, the single study with the larger patient cohort was included, or if more data were available on steroid response and biopsy results. Authors were contacted to avoid counting the same population twice. Six included studies were in French and translated into English. Studies in other languages did not meet inclusion criteria.

Risk of bias predominantly arose from lack of follow-up data. Quality of the papers were assessed using an adapted model [6], and all but two papers were found to be of reduced quality (Supplementary File). The model to assess quality was chosen for comparison to a similar study evaluating the epidemiology of chronic kidney disease in sub-Saharan Africa [6]. In total, 608 relevant published articles were found plus an additional 24 were added from review of references (Fig. 1). After excluding duplicates and all studies that did not meet the inclusion criteria, 81 records were included for analysis. Out of the 81 papers, 61 reported on steroid response and 63 on histology. One study had two separate study periods and was split into two studies [7]. A sensitivity analysis was performed by calendar year to assess temporal trends within each country as well as combined across the continent. Quartan malaria and hepatitis B infections, for example, have decreased in incidence. We performed the sensitivity analysis to determine if secondary causes of nephrotic syndrome have decreased accordingly. In the sensitivity analysis we compared studies with a study period before 1990 to those with study periods after 1990. To reduce publication bias, in the sensitivity analysis, we included only countries with studies reporting from both periods. There were five studies that crossed the 1990 threshold (eg. from 1985 to 1995) and these five studies were placed in the pre-1990 category with the predominate cases. Due to the small number of studies per country, we could not have smaller time windows to assess change except for Nigeria (Appendix Table 3 and 4).

Fig. 1.

PRISMA flow diagram displaying literature search method with included and excluded studies.

To determine the prevalence of steroid-sensitive nephrotic syndrome, steroid-resistant nephrotic syndrome, MCD, and FSGS, we calculated both the pooled prevalence with 95% confidence intervals (CI) and the median from each individual study then averaged by country. Steroid-resistance was defined by individual studies and a harmonized definition could not be used in this review. Calculations were performed to determine the total, using all included studies, as well as the total stratified by country and by region depicted in the tables. Similar calculations were performed for the sensitivity analysis. Figures were generated using the pooled calculation stratified by country along with the total. All analyses were calculated using STATA version 14·1.

Results

There were 81 papers from 17 countries out of the 54 African countries published between 1963 and 2020. Most African countries (n = 37, 69%) had no published data on childhood nephrotic syndrome. Majority of studies were from Nigeria (n = 34) [[7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40]], Egypt (n = 7) [[41], [42], [43], [44], [45], [46], [47]], and South Africa (n = 7) [[48], [49], [50], [51], [52], [53], [54]] and accounted for 59% of all studies (Fig. 2). The rest of the studies were from Senegal (n = 3) [[55], [56], [57]], Ghana (n = 3) [[58], [59], [60]], Gambia (n = 1) [61], Cote d'Ivoire (n = 2) [62,63], Zimbabwe (n = 1) [64], Namibia (n = 1) [65], Malawi (n = 2) [66,67], Tunisia (n = 4) [[68], [69], [70], [71]], Sudan (n = 4) [[72], [73], [74], [75]], Morocco (n = 3) [[76], [77], [78]], Libya (n = 4) [[79], [80], [81], [82]], Uganda (n = 3) [[83], [84], [85]], Ethiopia (n = 1) [86], and the Democratic Republic of Congo (n = 1) [87]. By study design, 62% were retrospective; 12% 20% and 6% were cross-sectional, prospective, and case-control, respectively. Mean age at diagnosis was 7.3 years with a male predominance of 62% (Table 1).

Fig. 2.

Published data on childhood nephrotic syndrome among individual African counties and the years of study.

Table 1.

Study population from 81 papers on childhood nephrotic syndrome in Africa.

	Country	Study quality (N studies)	% Retrospective	Children with nephrotic syndrome (n)	Age in years mean (SD)	% MaleA
Central Africa	Democratic Republic of Congo	Low (1)	100	62	7·5 (··)	53
West Africa	Côte d'Ivoire	Low (2)	100	173	6·9 (··)	63
	Gambia	Low (1)	100	39	6·6 (··)	69
	Ghana	Low (3)	67	229	8·6 (3·0)	65
	Nigeria	Medium (1)	51	2514	7·6 (1·4)	61
		Low (33)
	Senegal	Low (3)	67	173	8·1 (1·5)	66
North Africa	Egypt	Medium (1)	57	1119	6·4 (2·3)	64
		Low (6)
	Libya	Low (4)	100	413	4·9 (0·8)	62
	Morocco	Low (3)	100	171	12·2 (6·0)	56
	Sudan	Low (4)	75	776	6·2 (1·7)	67
	Tunisia	Low (4)	100	210	6·3 (1·1)	58
East Africa	Ethiopia	Low (1)	100	80	5·0 (··)	60
	Uganda	Low (3)	0	41	8·8 (2·1)	56
Southern Africa	Malawi	Low (2)	0	25	8·5 (··)	65
	Namibia	Low (1)	100	70	7·2 (··)	69
	South Africa	Low (7)	71	2201	5·5 (0·6)	62
	Zimbabwe	Low (1)	0	17	··	··
Total		−81	62	8313	7.3 (2.4)	62

(. .) Represent missing information or information not available.

^A% Males averaged across all studies for each country.

Steroid response

Among 61 studies from 15 countries, steroid-sensitive nephrotic syndrome accounted for 64% [95% CI: 63–66%] children, and steroid-resistant nephrotic syndrome accounted for 34% [95% CI: 33–35%]. There were 90 children (2%) with unknown or missing data. By region, pooled analysis demonstrated highest proportion of steroid-sensitive nephrotic syndrome in Central Africa (74%), however, the number of published reports differed by region. By country, Libya, Cote d'Ivoire, and Senegal had the highest proportions of steroid-sensitive nephrotic children between 80 and 95% (Table 2, Fig. 3). For steroid-resistant nephrotic syndrome by region, Northern Africa and Southern Africa had the highest and Central Africa the lowest proportions though the difference was only 10%. The highest proportions of steroid-resistant nephrotic syndrome were reported in Uganda (68%), Morocco (65%), and Egypt (54%). Only these three countries reported a higher prevalence of children with steroid-resistant than steroid-sensitive nephrotic syndrome.

Table 2.

Prevalence of steroid-sensitive nephrotic syndrome and steroid-resistant nephrotic syndrome among children by African country.

Region	Country	Given steroids	SSNSA	SRNSB	Prevalence SSNSA	Prevalence SRNSB
	(N studies)	(N)	(n)	(n)	Pooled (95% CI)F	Pooled (95% CI)F
					Median Individual (IQR)G	Median Individual (IQR)G
Central Africa	DRCC (1)	61	45	16	0·74 (0·61–0·84)	0·26 (0·16–0·39)
					0·74 (··)	0·26 (··)
	Total (1)	61	45	16	0·74 (0·61–0·84)	0·26 (0·16–0·39)
					0·74 (··)	0·26 (··)
West Africa	Côte d'Ivoire (2)	173	146	27	0·84 (0·78–0·89)	0·16 (0·11–0·22)
					0·84 (0·84–0·85)	0·16 (0·15–0·16)
	Gambia (0)	··	··	··	··	··
					··	··
	Ghana (3)	214D	73	55	0·34 (0·28–0·41)	0·26 (0·20–0·32)
					0·50 (0·29–0·70)	0·30 (0·21–0·50)
	Nigeria (24)	1560E	1001	555	0·64 (0·62–0·67)	0·36 (0·33–0·38)
					0·64 (0·41–0·83)	0·36 (0·17–0·59)
	Senegal (2)	148	118	30	0·80 (0·72–0·86)	0·20 (0·14–0·28)
					0·82 (0·77–0·87)	0·18 (0·13–0·23)
	Total (31)	2095	1338	667	0·64 (0·62–0·66)	0·32 (0·30–0·34)
					0·69 (0·46–0·84)	0·29 (0·16–0·50)
North Africa	Egypt (7)	910	415	495	0·46 (0·42–0·49)	0·54 (0·51–0·58)
					0·67 (0·37–0·75)	0·33 (0·25–0·63)
	Libya (4)	386	367	19	0·95 (0·92–0·97)	0·05 (0·03–0·08)
					0·97 (0·89–0·98)	0·03 (0·02–0·11)
	Morocco (1)	23	8	15	0·35 (0·16–0·57)	0·65 (0·43–0·84)
					0·35 (··)	0·65 (··)
	Sudan (3)	574	433	141	0·75 (0·72–0·79)	0·25 (0·21–0·28)
					0·72 (0·66–1·00)	0·28 (0–0·34)
	Tunisia (4)	210	130	80	0·62 (0·55–0·69)	0·38 (0·31–0·45)
					0·45 (0·06–0·84)	0·55 (0·16–0·94)
	Total (19)	2103	1353	750	0·64 (0·62–0·66)	0·36 (0·34–0·38)
					0·72 (0·37–0·88)	0·28 (0·12–0·63)
East Africa	Ethiopia (1)	80	61	19	0·76 (0·65–0·85)	0·24 (0·15–0·35)
					0·76 (··)	0·24 (··)
	Uganda (2)	25	8	17	0·32 (0·15–0·54)	0·68 (0·46–0·85)
					0·60 (0·19–1·00)	0·40 (0–0·81)
	Total (3)	105	69	36	0·66 (0·56–0·75)	0·34 (0·25–0·44)
					0·76 (0·19–1·00)	0·24 (0–0·81)
Southern Africa	Malawi (1)	10	7	3	0·70 (0·35–0·93)	0·30 (0·07–0·65)
					0·70 (··)	0·30 (··)
	Namibia (1)	22	12	10	0·55 (0·32–0·76)	0·46 (0·24–0·68)
					0·55 (··)	0·45 (··)
	South Africa (5)	1507	973	534	0·65 (0·62–0·67)	0·35 (0·33–0·38)
					0·56 (0·56–0·58)	0·44 (0·42–0·44)
	Zimbabwe (0)	··	··	··	··	··
					··	··
	Total (7)	1539	992	547	0·64 (0·62–0·67)	0·36 (0·33–0·38)
					0·56 (0·55–0·70)	0·44 (0·30–0·45)
Total	61	5903	3797	2016	0·64 (0·63–0·66)	0·34 (0·33–0·35)
					0·70 (0·48–0·82)	0·30 (0·18–0·50)

(. .) represent missing information or information not available.

^ASSNS = steroid-sensitive nephrotic syndrome.

^BSRNS = steroid-resistant nephrotic syndrome.

^CDRC=Democratic Republic of Congo.

^DThere were 86 children given steroids where the response (sensitive or resistant) is unknown.

^EThere were 4 children given steroids where the response (sensitive or resistant) is unknown.

^FThe pooled prevalence from all included studies within that country with 95% confidence intervals for each country/total.

^GMedian from each individual study averaged across the country with the interquartile range for each country/total.

Fig. 3.

Steroid response among children with nephrotic syndrome separated by African country calculated using pooled averages.

There were six countries that reported data from both pre- and post-1990 (Appendix Table 1). The overall pooled proportion of children with steroid-sensitive nephrotic syndrome increased slightly from 66% [95% CI: 63–88%] pre-1990 to 69% [95% CI: 66–71%] post-1990. Steroid-resistant nephrotic syndrome declined from 34% [95% CI: 32–37%] pre-1990 to 26% [95% CI: 24–28%] post-1990 (Appendix Table 1). Nigeria had a steady increase in steroid-sensitive and corresponding decline in steroid-resistant nephrotic syndrome. Proportions of steroid-sensitive nephrotic syndrome went from 51% before 1990, to 63% between 1990 and 2000, and 76% after 2000. Further stratification demonstrated a significant drop in reported steroid-resistant nephrotic syndrome from 49%, to 37% and then 24% during each respective time period.

Histology

When performing renal biopsy, centers may use light microscopy, electron microscopy, and immunofluorescence. Among the studies included, few reported performing all three; the majority only used light microscopy, and many performed no biopsy at all. Only seven countries reported the indications for renal biopsy. The predominant indication was steroid resistance, however other common indications included steroid dependence or frequently relapsing, and atypical nephrotic syndrome.

Renal biopsy histology was reported in 63 studies (78%) (Table 3, Fig. 4). The total proportion of minimal change and FSGS out of those biopsied with nephrotic syndrome was 38% [95% CI: 36–40%] and 24% [95% CI: 22–25%] respectively. The proportion of other histological types was 38% [95% CI: 36–40%]. By region, pooled analysis showed a significant difference between the highest proportion of minimal change histology in Central Africa (58%) compared to the lowest proportion in West Africa (15%). By country, the highest proportions of minimal change were found in Libya (88%), Morocco (69%), and the Democratic Republic of Congo (58%). By region, there were fewer differences between the highest proportion of FSGS in Central Africa (33%) compared to the lowest proportion of FSGS in East Africa (19%). By country, the highest proportions of FSGS were seen in Tunisia (64%), and Malawi (50%). Cote D'Ivoire reported only one child biopsied with FSGS pathology. Other histopathological types included, but were not limited to membranoproliferative glomerulonephritis, membranous glomerulonephritis, diffuse proliferative glomerulonephritis and more (Supplementary File). West Africa had the highest proportion of other histological types (62%) compared to only 8% in Central Africa. In Zimbabwe, secondary forms represent 82% of biopsies, 76% in Namibia and 65% in Nigeria.

Table 3.

Prevalence of minimal change disease, focal segmental glomerulosclerosis and other histological types among children by African country.

	Country	Biopsied	MCDA	FSGSB	OtherC	Prevalence MCDA	Prevalence FSGSB	Prevalence OtherC
	(N studies)	(N)	(n)	(n)	(n)	Pooled (95% CI)L	Pooled (95% CI)L	Pooled (95% CI)L
						Median Individual (IQR)M	Median Individual (IQR)M	Median Individual (IQR)M
Central Africa	DRCD (1)	12	7	4	1	0·58 (0·28–0·85)	0·33 (0·10–0·65)	0·08 (0–0·38)
						0·58 (··)	0·33 (··)	0·08 (··)
	Total (1)	12	7	4	1	0·58 (0·28–0·85)	0·33 (0·10–0·65)	0·08 (0–0·38)
						0·58 (··)	0·33 (··)	0·08 (··)
West Africa	Côte d'Ivoire (1)	1	0	1	0	0 (0–0·98)	1·00 (0·02–1)	0 (0–0·98)
						0 (··)	1 (··)	0 (··)
	Gambia (0)	··	··	··	··	··	··	··
						··	··	··
	Ghana (2)	38	16	11	11	0·42 (0·26–0·59)	0·29 (0·15–0·46)	0·29 (0·15–0·46)
						0·36 (0·15–0·56)	0·40 (0·04–0·77)	0·24 (0·08–0·4)
	Nigeria (25)	683E	91	122	445	0·13 (0·11–0·16)	0·21 (0·18–0·24)	0·65 (0·61–0·69)
						0·17 (0·04–0·23)	0·30 (0·04–0·52)	0·56 (0·20–0·67)
	Senegal (3)	48	11	13	24	0·23 (0·12–0·37)	0·27 (0·15–0·42)	0·50 (0·35–0·65)
						0·29 (0–0·53)	0·24 (0·21–0·57)	0·24 (0·14–0·79)
	Total (31)	770F	118	147	480	0·15 (0·13–0·18)	0·22 (0·19–0·25)	0·62 (0·59–0·66)
						0·17 (0–0·27)	0·30 (0·05–0·57)	0·40 (0·14–0·67)
North Africa	Egypt (7)	1020	521	196	303	0·51 (0·48–0·54)	0·19 (0·17–0·22)	0·30 (0·27–0·33)
						0·33 (0·29–0·54)	0·30 (0·16–0·33)	0·29 (0–0·40)
	Libya (1)	17	15	0	2	0·88 (0·64–0·99)	0 (0–0·20)	0·12 (0.01–0·36)
						0·88 (··)	0 (··)	0·12 (··)
	Morocco (3)	240G	72	36	20	0·69 (0·59–0·77)	0·15 (0·11–0·20)	0·19 (0·12–0·28)
						0·71 (0·63–0·79)	0·13 (0·12–0·17)	0·16 (0·09–0·24)
	Sudan (4)	308	138	64	106	0·45 (0·39–0·51)	0·21 (0·16–0·26)	0·34 (0·29–0·40)
						0·34 (0·17–0·45)	0·26 (0·14–0·46)	0·38 (0·29–0·50)
	Tunisia (4)	68	25	33	10	0·26 (0·17–0·35)	0·64 (0·54–0·74)	0·10 (0·05–0·18)
						0·33 (0·13–0·45)	0·54 (0·43–0·78)	0·11 (0·02–0·18)
	Total (19)	1653H	771	359	441	0·50 (0·47–0·52)	0·21 (0·19–0·23)	0·28 (0·26–0·31)
						0·38 (0·29–0·54)	0·28 (0·13–0·53)	0·22 (0·09–0·36)
East Africa	Ethiopia (0)	··	··	··	··	··	··	··
						··	··	··
	Uganda (2)	36	14	7	15	0·39 (0·23–0·57)	0·19 (0·08–0·36)	0·42 (0·26–0·59)
						0·23 (0–0·45)	0·11 (0–0·23)	0·66 (0·32–1·00)
	Total (2)	36	14	7	15	0·39 (0·23–0·57)	0·19 (0·08–0·36)	0·42 (0·26–0·59)
						0·23 (0–0·45)	0·11 (0–0·23)	0·66 (0·32–1·00)
Southern Africa	Malawi (1)	10	0	5	5	0 (0–0·31)	0·50 (0·19–0·81)	0·50 (0·19–0·81)
						0 (··)	0·50 (··)	0·50 (··)
	Namibia (1)	68	5	11	52	0·07 (0·02–0·16)	0·16 (0·08–0·27)	0·76 (0·65–0·86)
						0·07 (··)	0·16 (··)	0·76 (··)
	South Africa (7)	1280I	411	296	402	0·39 (0·36–0·42)	0·28 (0·26–0·31)	0·31 (0·29–0·34)
						0·43 (0·36–0·47)	0·32 (0·04–0·43)	0·28 (0·04–0·54)
	Zimbabwe (1)	17	1	2	14	0·06 (0–0·29)	0·12 (0·01–0·36)	0·82 (0·57–0·96)
						0·06 (··)	0·12 (··)	0·82 (··)
	Total (10)	1375J	417	314	473	0·37 (0·34–0·39)	0·28 (0·25–0·30)	0·34 (0·32–0·37)
						0·36 (0·07–0·46)	0·28 (0·12–0·43)	0·43 (0·27–0·56)
Total	63	3711K	1327	831	1410	0·38 (0·36–0·40)	0·24 (0·22–0·25)	0·38 (0·36–0·40)
						0·27 (0·08–0·48)	0·28 (0·11–0·50)	0·32 (0·14–0·59)

(. .) Represent missing information or information not available.

^AMCD = minimal change disease.

^BFSGS = focal segmental glomerulosclerosis.

^COther includes any of the following: membranous nephropathy, tropical extramembranous, mesangial proliferative glomerulonephritis, cyclosporine toxicity, likely FSGS, membranoproliferative glomerulonephritis, amyloidosis, diffuse mesangial proliferation, diffuse segmental glomerulonephritis, diffuse proliferative glomerulonephritis, IgA, dense deposit disease, IgM, post infectious glomerulonephritis, lupus nephritis, Alport's syndrome, tropical immune complex glomerulonephritis, immune complex glomerulonephritis and secondary FSGS, mesangiocapillary glomerulonephritis, diffuse global glomerulonephritis, diffuse mesangial hypercellularity, quartan malaria nephropathy, proliferative, extra capillary proliferation, mild mesangial hypercellularity, post-streptococcus glomerulonephritis, endocapillary glomerulonephritis, focal proliferative secondary to sickle cell, chronic glomerulonephritis, crescentic glomerulonephritis, congenital.

^DDRC=Democratic Republic of Congo.

^E# biopsied (denominator) for MCD and Other is 683; # biopsied (denominator) for FSGS is 585. There were 25 children with unknown data.

^F# biopsied (denominator) for MCD and Other is 770; # biopsied (denominator) for FSGS is 672.

^G# biopsied (denominator) for MCD and Other is 105; # biopsied (denominator) for FSGS is 240. There were 112 children with unknown data.

^H# biopsied (denominator) for MCD and Other is 1518; # biopsied (denominator) for FSGS is 1653.

^I# biopsied (denominator) for MCD and FSGS is 1046; # biopsied (denominator) for Other is 1280. There were 171 children with unknown data.

^J# biopsied (denominator) for MCD and FSGS is 1141; # biopsied (denominator) for Other is 1375.

^K# biopsied (denominator) for MCD is 3477; # biopsied (denominator) for FSGS is 3514; # biopsied (denominator) for Other is 3711.

^LThe pooled prevalence from all included studies within that country with 95% confidence intervals for each country/total.

^MThe median from each individual study averaged across the country with the interquartile range for each country/total.

Fig. 4.

Proportions of focal segmental glomerulosclerosis (FSGS) minimal change disease (MCD) and other histological types out of those biopsied from 1960 to 2018 in African countries calculated using pooled averages.

^aCote D'Ivoire had one study with one biopsy which showed Focal Segmental Glomerulosclerosis and was not included in this figure.

^bNumber biopsied for FSGS = 585.

^cNumber biopsied for MCD and other = 105.

^dNumber biopsied for MCD and FSGS = 1046.

^eNumber biopsied for Total MCD = 3477 and Total FSGS = 3514.

^fOther includes any of the following: membranous nephropathy, tropical extramembranous, mesangial proliferative glomerulonephritis, cyclosporine toxicity, likely FSGS, membranoproliferative glomerulonephritis, amyloidosis, diffuse mesangial proliferation, diffuse segmental glomerulonephritis, diffuse proliferative glomerulonephritis, IgA, dense deposit disease, IgM, post infectious glomerulonephritis, lupus nephritis, Alport's syndrome, tropical immune complex glomerulonephritis, immune complex glomerulonephritis and secondary FSGS, mesangiocapillary glomerulonephritis, diffuse global glomerulonephritis, diffuse mesangial hypercellularity, quartan malaria nephropathy, proliferative, extra capillary proliferation, mild mesangial hypercellularity, post-streptococcus glomerulonephritis, endocapillary glomerulonephritis, focal proliferative secondary to sickle cell, chronic glomerulonephritis, crescentic glomerulonephritis, congenital.

Only five countries reported histology data from both pre- and post-1990 (Appendix Table 2). After comparison of trends before and after 1990, the overall pooled proportion of children with minimal change was 24% [95% CI: 21–27%] pre-1990 and 30% [95% CI: 26–34%] post-1990. For FSGS, pooled proportion was 13% [95% CI: 10–15%] pre-1990, rising to 41% [95% CI: 37–45%] post-1990. For other histopathological types pooled proportion was 54% [95% CI: 51–57%] pre-1990, declining to 29% [95% CI: 25–33%] post-1990.

Discussion

In 2010, the World Health Organization reported that 2·3–7·1 million people died from end-stage kidney disease globally [88]. In Africa, the burden of chronic kidney disease is 14% among one billion people on the continent ranging in age from 35 to 45.9 years [6]. If kidney disease is treated and prevented early in childhood, progression to end-stage kidney disease can be delayed and potentially cured. Childhood nephrotic syndrome in Africa is one of the leading causes of chronic kidney disease [2,88]. In this systematic review on childhood nephrotic syndrome, only a third of African countries reported on this treatable childhood form of kidney disease. The majority of countries with data are listed in the top 20 of nominal gross domestic product of Africa, which reflects differential access to care and nephrology training across the continent [89].

There are notable differences in the epidemiology of childhood nephrotic syndrome in Africa compared to the described predominantly European populations from Europe and North America. We report a mean age at diagnosis of 7.3 years old, which is three to four years older than reported in those of European and South Asian ancestry [90]. The older age reported here may be due to the fact that nephrotic syndrome in Africa is primarily managed by general pediatricians. A lack of nephrologists across the continent may result in delayed diagnosis. Early recognition of childhood nephrotic syndrome is important and can be mistaken for protein energy malnutrition [91]. Similar to other studies, there is a male predominance in African children.

Steroid-sensitive nephrotic syndrome is the most common form of childhood nephrotic syndrome in Africa. Variations in steroid response are modest across the continent with slight regional differences, which may be due to publication bias, with few papers from any country in a given region, or selection bias of those from regional centers and not a sampling strategy of all children at risk. In Western Africa, Nigeria has the vast majority of published papers compared to other countries, thus, trends could be detected, however, may not be generalizable to the rest of the continent. There is an increased need for access and early use of steroids across the continent in order to treat these children.

While the majority of the children had steroid-sensitive nephrotic syndrome, 34% had steroid-resistant nephrotic syndrome, a much higher rate than reported among European Americans (20%) [5]. The studies, however, did not use a uniform definition of steroid resistance, which could lead to misclassification. Nonetheless, treatment with steroid-sparing medications such as cyclosporine can prevent progressive kidney disease but requires close monitoring to prevent long-term kidney damage [92]. Access and cost of second-line medications with frequent laboratory tests can be prohibitive since most countries lack health coverage by either government or private health insurance. If children with steroid-resistant nephrotic syndrome are left untreated, 50% are at risk of progressing to end-stage kidney disease within five years [1]. Interestingly, temporal trends demonstrate a decreasing proportion from 34 to 26% of steroid resistance over the decades. A lower prevalence over time suggests improvement in access to consultative nephrology care and pathology support in addition to possibly an improved environment with less infections and improved generalized health status in a region. Alternatively, it could still represent a reporting bias given the lack of a cohesive definition for steroid-resistant nephrotic syndrome. Once the recently published paper, “Management of idiopathic childhood nephrotic syndrome in Sub-Saharan Africa: Ibadan consensus statement” is implemented more widely in Africa, incidence of steroid-resistant nephrotic syndrome will need to be studied prospectively [93].

In this review, African children were biopsied typically due to a complex clinical course (eg. steroid resistance or atypical symptoms). In recent years, the trend shows an increase of minimal change and FSGS (24% to 30% and 13% to 41% respectively), while other histological types have decreased (54% to 29%). This may reflect better biopsy technique and more trained pathologists in recent years accurately describing the biopsy samples. This highlights the importance of understanding the epidemiology of nephrotic syndrome in Africa as the availability of specialized diagnostic testing of kidney tissue and serologic tests are necessary. Many countries in Africa do not have easy access to pathology unless paid for by the hospital or family funds.

The proportion of African children with minimal change is far lower than other biopsy cohorts from the International Study of Kidney Disease (ISKDC) with 70–80% [94]. These studies included children predominantly of European origin and all children with nephrotic syndrome were biopsied. Based on the report from ISKDC, management worldwide has presumed minimal change pathology typically for all children ages of two to twelve and a kidney biopsy is not indicated. More recently, a biopsy derived cohort (NEPTUNE) with both adult and pediatric participants reported that among Black participants in the US, 24% had minimal change, 35% had FSGS and 31% had other histopathologies [95]. This was a highly selected group from tertiary pediatric centers that conducted a biopsy for clinical indications. These two biopsy series based in Europe and in the US are in direct contrast to the findings in Africa, however, the selection of the study participants varied in all three studies. Given there were no children of African ancestry in ISKDC and the findings from NEPTUNE among African Americans, it may be important to consider whether guidelines on the timing of biopsy among African children may differ. Additionally, if pathology or more detailed serological evaluation is available results may tailor treatments to prevent progressive kidney disease. It is important for physicians in Africa to consider the need for biopsy, if attainable in the region, and not make the presumptive diagnosis of minimal change histology. We recognize, however, that the cost of serology testing and kidney biopsy may not be available to all patients. This is an important consideration for hospitals providing charitable care and evolving health care plans in Africa. Despite the higher prevalence (42–74%) of FSGS described among African Americans in the US, we report a relatively low prevalence of 24% overall [3]. There were slight regional differences, which could be due to varying genetic risks such as the gene APOL1, which was originally identified with end-stage renal disease [96]. With the higher frequency of APOL1 high-risk variants reported in Western Africa, the lower risk of FSGS is not clear [97]. It is important to note that APOL1 is associated with kidney disease progression and not specific to FSGS. Studies from Western Africa constitute the largest number with over 700 children and the findings are consistent among individual publications. This may explain why West Africa reported higher rates of steroid resistance and lower rates of FSGS. Presumably, some of the steroid resistant children would have shown FSGS histology had they received a kidney biopsy.

In Africa, children with secondary diseases, including membranous and membranoproliferative glomerulonephritis, accounted for almost 40% with a wide range from 8 to 82% in a given country. Given temporal trends of improved pathology, use of immunohistochemistry, and electron microscopy as well as serology, accuracy will improve our understanding of secondary causes of nephrotic syndrome. In recent years, the proportion of secondary cases has decreased from 54% to 29%, although, the prevalence remains higher than in children of European descent [90]. Unlike studies from high income countries, children in African countries have a higher burden of kidney diseases secondary to infections. Among the 38% of other histological types, infection was a common cause, including malaria [[30], [31], [32], [33],98]. As the rate of quartan malaria in Africa decreased, so did the rate of secondary causes of nephrotic syndrome, especially in Nigeria, with a decrease from 85% to 42% after 1990. Hepatitis B virus, human immunodeficiency virus, or sickle cell disease, and systemic lupus erythematosus are also more commonly diagnosed in recent years [99,[100], [101], [102]]. Lupus, often thought of as rare in Africa, may be increasing in prevalence [103]. The high proportions of secondary causes of nephrotic syndrome may warrant routine biopsy and serological evaluation at the time of presentation of nephrotic syndrome, which will alter treatment options to prevent progression to end-stage renal disease in African children.

In a recent systematic review on chronic kidney disease in sub-Saharan Africa, kidney function measures were inconsistently collected, preventing an assessment of health burden on the continent [6]. This is also evident among studies reviewed for childhood nephrotic syndrome. We found few published reports across the African continent overall, studies with small sample sizes, inconsistencies in definitions especially steroid resistance as well as insufficient follow-up data on treatment, outcomes, and kidney function [61]. Prospective studies are needed to confirm these findings. In many African countries, children may also be treated with alternative medicine as reported in the study from the Democratic Republic of Congo where about 10% of patients consulted traditional healers [87]. There is also publication bias as some regions do not have pediatric nephrologists, limited infrastructure or language barriers to support publication. Moreover, there is a lack of trained pediatricians and pediatric nephrologists in low resource settings throughout Africa. Adult nephrologists often care for teenagers in most countries and may not use child specific protocols for treatment [93]. If clinicians are not aware of treatment recommendations, this could result in inadequate treatment of children leading to increased risk for chronic kidney disease, morbidity, and mortality.

Patterns of nephrotic syndrome in African children vary compared to those from European or North American studies. The majority have steroid-sensitive nephrotic syndrome, and biopsy findings highlight that minimal change and secondary causes of nephrotic syndrome are equally common accounting for 76% and FSGS, although less common occurs in a quarter of cases. Evaluation by a nephrologist, with appropriate serological evaluation and kidney biopsy are important aspects of management of childhood nephrotic syndrome in Africa. The majority of children with nephrotic syndrome are responsive to treatments, and with improved recognition and access to treatment, progression to chronic kidney disease can be prevented.

Research in context

Evidence before this study:

Previous reports on childhood nephrotic syndrome across the African continent are scarce. For this systematic review, we searched MEDLINE, Embase, African Journals Online, and WHO Global Health Library from January 1, 1946 to July 1st 2020 for studies in English or French reporting on childhood nephrotic syndrome. Search terms included “nephrotic syndrome” or “edema hypertension proteinuria syndrome” or “oedema hypertension proteinuria syndrome” as well as “pediatric” “adolescent” and every country within the African continent. Studies were excluded if they reported only on adults, were located outside Africa, or only reported secondary causes of nephrotic syndrome. A prior systematic review from 2017 on childhood nephrotic syndrome only included countries in Tropical Africa and data from the entire continent is lacking as well as pooled analyses by country and region.

Added value of this study:

This systematic review provides an overview of the trends of childhood nephrotic syndrome in Africa. We found that 31% of African countries reported on childhood nephrotic syndrome, the most common form of childhood kidney disease, demonstrating a lack of data across the continent. The mean age at presentation was 7.3 years, higher than reports from the United States and Europe. We found a predominantly male population similar to global reports. Steroid sensitive nephrotic syndrome is more common than steroid resistant disease among children in Africa with proportions of 64% and 34% respectively. Temporal trends demonstrate an increasing prevalence of steroid sensitive nephrotic syndrome and declining prevalence of steroid resistant nephrotic syndrome. Clinical indications for biopsy demonstrated that minimal change pathology account for almost 40%, which is as common as secondary causes of nephrotic syndrome and 24% have FSGS. Histology patterns differ from studies in the United States and Europe with predominantly European populations.

Implications of all the available evidence:

This systematic review highlights the need for more prospective data on childhood nephrotic syndrome throughout the African continent as the patterns may lead to differences in diagnosis and treatments. Increased trained pediatric nephrologists are needed to diagnose and treat the disease. Most children have steroid sensitive disease, thus, increased early recognition and treatment with steroids is necessary. For the children who are steroid resistant, second line medication is warranted which may reduce the progression to chronic kidney disease. Childhood nephrotic syndrome in Africa has a different disease pattern; thus, global guidelines which presume minimal change disease and do not suggest biopsy at diagnosis may not be suitable among countries in Africa. We suggest the potential need for biopsy and serological evaluation at time of diagnosis in order to diagnose and treat the large proportion of secondary causes seen across the continent. Childhood nephrotic syndrome is treatable and with the appropriate management throughout Africa, the progression to chronic kidney disease can be delayed and prevented.

Credit author statement

RSP: Conceptualization, Methodology, Writing – Original Draft, Visualization, Supervision, Funding acquisition.

RW: Conceptualization, Methodology, Formal analysis, Investigation, Writing – Original Draft, Visualization.

JVR: Conceptualization, Methodology, Investigation, Writing – Original Draft, Visualization, Supervision.

TB: Formal analysis, Investigation, Writing – Original Draft, Visualization.

JK, DN, RG, TOI, HUO, FF, IU, AUS, CE, AB, YR, TOO, UM, AEA, VB, OA, CO, GG, SA, MAM, EAA, PR, AOA, ADA: Investigation, Writing – Review and Editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary material 1

Supplementary material 2