Burden of chronic kidney disease on the African continent: a systematic review and meta-analysis

Burden of chronic kidney disease on the African continent: a systematic review and meta-analysis Background: Accurate contemporary data on the burden of Chronic Kidney Disease (CKD) on the African continent are lacking. We determined the prevalence of CKD in adult populations living in Africa, and variations by stage, gender, estimated Glomerular Filtration Rate (eGFR) equation, and residence. Methods: For this systematic review, we searched multiple electronic databases for original studies on CKD prevalence reported from January 1, 2000 to December 31, 2016. Two reviewers independently undertook quality assessment and data extraction. We stabilized the variance of study-specific estimates with the Freeman-Turkey single arcsine transformation and pooled the data using a random effects meta-analysis models. Results: A total of 98 studies involving 98,432 individuals were included in the final meta-analysis. The overall prevalence was 15.8% (95% CI 12.1–19.9) for CKD stages 1–5 and 4.6% (3.3–6.1) for CKD stages 3–5 in the general population. Equivalent figures were greater at 32.3% (23.4–41.8) and 13.3% (10.7–16.0) in high-risk populations (people with hypertension, diabetes, HIV). CKD prevalence was higher in studies based on the Cockcroft-Gault formula than MDRD or CKD-EPI equations; and in studies from sub-Saharan Africa compared with those from North Africa (17.7, 95% CI 13.7–22.1 vs 6.1, 95% CI 3.6–9.3, p < 0.001). There was substantial heterogeneity across studies (all I > 90%) and no evidence of publication bias in main analyses. Conclusion: CKD is highly prevalent across Africa, inviting efforts into prevention, early detection and control of CKD in adults living on the African continent which is particularly important in a resource limited environment. Trial Registration: Prospero Registration ID: CRD42017054445. Keywords: Chronic kidney disease, Prevalence, Systematic review, Meta-analysis, Africa Background of communicable and non-communicable diseases, in Chronic kidney disease (CKD) is a leading cause of mor- part driven by the adoption of western lifestyles, changes bidity and mortality in both developed and developing in the built environment, and the rapid urbanization [5]. countries, with an estimated 10% of the population This dual burden has led to a consequential rise in the worldwide having CKD in 2015 [1, 2]. Studies have con- number of people affected by CKD on the African con- sistently shown that African descendants are at in- tinent [6]. creased risk for CKD occurrence and progression to Given the constant rise in its risk factors in Africa [7, 8], end-stage renal disease (ESRD) [3, 4]. Many African CKD is increasingly recognized as a major public health countries are currently undergoing rapid epidemiological threat, against a background of limited access to renal re- transitions and are confronted with the double burden placement therapy (RRT) [6]. Hence, in Africa, prevention and early detection of CKD in order to slow its progres- sion are of paramount importance. For this purpose, a bet- * Correspondence: jechouffotcheugui@bwh.harvard.edu ter understanding of the current prevalence of CKD in Division of Endocrinology, Diabetes, and Hypertension, Brigham and Africa is urgently needed. Although the number of reports Women’s Hospital, Harvard Medical School, 221 Longwood Avenue, Boston, MA 02115, USA on CKD prevalence across Africa has increased in recent Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Kaze et al. BMC Nephrology (2018) 19:125 Page 2 of 11 years, accurate data on its exact magnitude on the contin- according to the NKF KDOQI (National Kidney Founda- ent are still lacking [6]. The only systematic review of tion Kidney disease outcomes quality initiative or KDIGO CKD prevalence in Africa was limited to sub-Saharan (Kidney Disease: Improving Global Outcomes) guidelines countries, included studies published between 1962 and [13, 14]. We excluded studies in participants selected 2011, and highlighted the inability to make definitive in- based on the presence or absence of kidney disease, stud- ferences due to the poor quality of included studies [9]. ies limited to pregnant women. For multiple surveys con- We conducted a systematic review and meta-analysis of ducted in different countries or at different calendar years the contemporary evidence on CKD prevalence in adults and reported within the same article, each survey was living on the African continent, in order to establish base- accounted for separately, when it was possible to disaggre- line figures against which future trends can be monitored. gate the data by country. Methods Assessment of the methodological quality of included We used a prospective protocol (PROSPERO studies CRD42017054445) [10] to perform this review according Two reviewers (ADK & TI) independently assessed to the Preferred Reporting Items for Systematic Reviews study quality, with disagreements being resolved by con- and Meta-analyses (PRISMA) guidelines [11]. sensus or by consulting a third investigator (JBE). We used the ten-item rating checklist developed by Hoy et Search strategy and selection criteria al. that assesses sampling, the sampling technique and Identification of relevant studies size, outcome measurement, response rate, and statis- We performed comprehensive electronic searches of tical reporting [15]. Each item was assigned a score of 1 major databases including PubMed/Medline and Embase (yes) or 0 (no), and scores were summed across items to using an African search filter developed by Eisinga and generate an overall quality score ranging from 0 to 10. colleagues [12], to identify relevant studies published on Each study was rated as being of low, moderate, or high CKD on the African continent between January 1, 2000 methodological quality depending on the number of and December 31, 2016, without language restriction questions answered as “yes (low risk of bias)”. Studies of (Additional file 1: pp. 2–3). The searches were restricted high quality had scores higher than 8, moderate a score to the post-2000 era in order to provide the most con- of 6–8, and low a score of 5 or lower [15]. temporary estimate of CKD prevalence in Africa. Esti- mates from previous studies are also likely based on Data extraction older definitions of CKD and therefore not directly com- The same reviewers independently selected studies and parable to more recent studies. Additionally, we traced extracted relevant data using a pre-conceived extraction the citations of identified articles via the ISI Web of form. The information extracted included 1) author de- knowledge, and scanned the reference lists of review pa- tails (names and year of publication); 2) study character- pers and conference proceedings. We also searched rele- istics (country, design, setting, data source, sampling vant African journals, the World Health Organization method, sample size, data collection period, response (WHO) Global Health Library databases (which include rate); 3) Participants’ characteristics (age, gender, hyper- the African Index Medicus, WHO Library Information tension status, diabetes status, HIV status, HAART treat- System, and Scientific Electronic Library Online). ment); and 4) CKD characteristics (CKD diagnostic criteria, eGFR equation, proteinuria assessment method, Selection of included studies prevalence, number of participants tested and diagnosed Two investigators (ADK and TI) independently reviewed with CKD overall and by subgroups of interest). the articles by title, abstract, and full-text where relevant for inclusion. Disagreements were resolved by consensus Statistical analysis or by consulting a third investigator (JBE). To be included For each study, the unadjusted prevalence of CKD and in this review, primary studies had to be population or standard errors were calculated (number of cases/sample hospital- based, randomized controlled trial (RCT) or size) based on the information on crude numerators and non-randomized studies including cross-sectional, cohort, denominators provided in individual studies. CKD stages and case-cohort studies that report prevalence or enough 1–5 was defined as kidney damage (urinary dipstick ab- data to calculate prevalence of CKD in adults (aged 18 normalities) or eGFR < 60 ml/min/1.73m , while CKD and above) from sub-Saharan and North African coun- stages 3–5 was defined as eGFR < 60 ml/min/1.73m . tries. For cohort studies or RCT, we included data from When eGFR from multiple equations was reported, we the baseline evaluation. CKD had to have been defined as preferred the Chronic Kidney Disease Epidemiology the presence of kidney damage and/or estimated Glom- Collaboration (CKD-EPI) equation, then the Modifica- erular filtration rate (eGFR) < 60 mL/min/1.73 m , tion of Diet in Renal Disease (MDRD), and lastly the Kaze et al. BMC Nephrology (2018) 19:125 Page 3 of 11 Cockcroft-Gault formula in the main analyses. We used (central, eastern, northern, southern and western Africa), the DerSimonian-Laird random-effects models to gener- sample size, and year of publication. Comparisons be- ate the pooled prevalence of CKD according to each tween subgroups were performed using the Q-test based diagnostic criterion. The random effects model was on the Analysis of the Variance (ANOVA). Publication chosen in anticipation of substantial variations in CKD bias was evaluated using funnel plots supplemented by prevalence estimates across the included studies. To formal statistical assessment using the Egger’s test [19]. minimize the effect of studies with extremely small or All analyses were performed using Stata software (Stata extremely large prevalence on the overall estimate, we Corp V.14, Texas, USA). stabilized the variance of the study-specific prevalence estimates with the Freeman-Tukey single arcsine trans- Results formation before pooling the data [16]. We examined The review process prevalence by disease-specific populations (general Fig. 1 summarizes the study selection process. In total, population, HIV, Hypertension, diabetes mellitus), re- 1429 records were identified via databases searches. gion, method of kidney disease assessment, setting, date. After removing duplicates, we scanned the titles and ab- We assessed inter-rater agreement for inclusion and stracts of 1388 studies, of which 259 were selected for quality assessment using Cohen’s kappa (κ) coefficient. full-text review. Of these, 98 met the inclusion criteria We assessed heterogeneity between studies using and were retained in the final review (Fig. 1). Inter-rater Cochran’s Q statistic, H and the I statistics [17, 18], agreement for inclusion was excellent (κ = 0.965). which estimate the percentage of total variation across studies due to true between-study difference rather than Methodological quality of included studies chance, with I values of 25, 50 and 75% representing A total of 19 studies were deemed to be of high meth- low, medium and high heterogeneity, respectively. We odological quality, while 78 were categorized as being of explored sources of heterogeneity through a restriction moderate methodological quality; two studies were con- of analyses to subgroups defined by geographical area sidered to be of low quality and were therefore excluded 1429 records identified through database searches 746 from Medline 635 from Embase 48 from references and other sources 1388 records screened after duplicates removed 1129 records excluded 259 full-text articles assessed for eligibility 161 full-text articles excluded 93 no data on outcome 29 no prevalence data 22 no original research 15 duplicates 2 studies with poor quality rating 98 studies included in synthesis Fig. 1 Selection of articles for inclusion in the systematic review Included Eligibility Screening Identification Kaze et al. BMC Nephrology (2018) 19:125 Page 4 of 11 from the analysis (Additional file 1: Table S1, pp. 5–8). Prevalence of CKD in the general population Inter-rater agreement for quality assessment was excel- Fig. 2 shows the prevalence of CKD in the general popu- lent (κ = 0.910). lation by region. The overall prevalence in the general population was 15.8% (95% CI 12.1–19.9, n = 23,825, 22 studies, Fig. 3) for CKD stages 1 to 5, and 4.6% (95% CI Characteristics of included studies 3.3–6.1, n = 25,929, 27 studies, Fig. 4) for CKD stages 3 The characteristics of the included studies are summa- to 5. The overall prevalence in the general population rized in Additional file 1: Table S2 (pp 9–13). Sixty four was significantly higher in sub-Saharan Africa compared (65.3%) articles were published between 2012 and 2016. to north Africa for both of CKD stages 1–5 (17.7, 95%CI They originated from the five African sub-regions, with 13.7–22.1; 10,028 participants vs 6.1, 95% CI 3.6–9.3, the western sub-region being the most represented (37 13,797 participants, p < 0.001, Additional file 1: Table S3, studies, 37.8%) and the northern being the least repre- pp. 14–15) and CKD stages 3–5 (4.8, 95% CI 3.2–6.6, sented (six studies, 6.1%). Of the 54 African countries, n = 15,132 vs 2.6, 95% CI 2.3–2.9, n = 10,797; p =0.004, 22 (40.7%) were represented in this systematic review in- Additional file 1:Table S4, pp. 15–16). cluding Nigeria (26 studies); South Africa, Ghana (nine studies each); Tanzania (eight studies); Cameroon, Prevalence of CKD in high-risk populations Democratic Republic of Congo (six studies each); The prevalence of CKD stages 1 to 5 in the high-risk Malawi (five studies); Uganda, Zimbabwe, Egypt, Kenya populations was 32.3% (95% CI 23.4–41.8, n = 5056; 21 (three studies each); Congo, Sudan, Ethiopia, Senegal, contributions, Fig. 5) overall, 27.3% in participants with Zambia (two studies each); Burundi, Morocco, Rwanda, HIV (95% CI 17.0–38.9, n = 2007; 10 contributions, Add- Seychelles (one study each). One study presented com- itional file 1: Figure S3, pp. 23), 35.6% in participants bined data from two surveys conducted in Zimbabwe with hypertension (95% CI 27.9–43.7, n = 2199, 6 contri- and Uganda, another study presented combined data butions, Additional file 1: Figure S4, pp. 24), and 32.6% from seven sub-Saharan countries (Cameroon, Ivory (95% CI 0.3–82.3, 778 participants, four contributions, Coast Kenya, Mozambique, South Africa, Uganda, and Additional file 1: Figure S5, pp. 25) in participants with Zambia). Of the included studies, 97 were local studies, diabetes (Additional file 1: Table S5, pp. 16–17). The mainly conducted in urban settings, while only one prevalence of CKD stages 3 to 5 in the high-risk popula- study had national coverage [20]. With regards to the tions was 13.3% (95% CI 10.7–16.0, 52,353 participants; study sites, the majority were solely hospital- or 50 contributions, Fig. 6) overall, 9.1% (95% CI 6.6–11.9, clinic-based (59 studies, 60.2%), with only 39 studies be- n = 44,239, 28 contributions, Additional file 1: Figure S6, ing community-based. pp. 26) in participants with HIV, 17.9% (95% CI 10.9– Data from 98,432 participants were included, with a 26.1, n = 2971, 11 contributions, Additional file 1: Figure median age of 43 years (25th–75th percentiles: 36.6 to S7, pp. 27) in participants with hypertension, and 51.4). Fifty four (55.1%) studies examined participants 22.0% (95% CI 16.1–28.6, n = 5071, 10 contributions, thought be at high risk of CKD (people with hyperten- Additional file 1: Figure S8, pp. 28) in participants sion, diabetes, HIV, or sickle cell disease), while 33 with diabetes (Additional file 1: Table S6, pp. 17–18). (33.7%) were conducted in the general population or subjects not known to be at risk of CKD, and 11 (11.2%) Prevalence of CKD by eGFR estimating equation studies in both. The included studies applied various es- In general populations, the prevalence of CKD stages 1–5 timators of GFR. Fifty eight applied a single equation; was 13.7% (95% CI 10.2–17.6, n = 23,825, 14 contribu- the most frequently used being the Modification of Diet tions) according to the MDRD equation, 21.3% (95% CI in Renal Disease (MDRD) eq. (32 studies), followed by 9.9–35.5 n = 5316, 8 contributions) according to the the Cockcroft-Gault formula (17 studies), the Chronic Cockcroft-Gault formula, and 19.5% (95% CI 13.8–25.9, Kidney Disease Epidemiology Collaboration (CKD-EPI) n = 4854, 5 contributions) when using the CKD-EPI formula (7 studies), and the Cystatin C equation (two formula (Additional file 1:FigureS1, pp. 19–20). The studies). Of the 22 studies that compared eGFR using prevalence of CKD stages 3–5 was 4.5% (95% CI 3.0–6.3, two or more equations, Cockcroft-Gault and MDRD n = 22,969, 21 contributions) by the MDRD equation, equations were used in 11 studies, CKD-EPI and 11.8% (95% CI 7.3–17.1, n = 4927, 11 contributions) ac- MDRD used in 3 studies, and the three formulas cording to the Cockcroft-Gault formula, and 5.1% (95% CI (Cockcroft-Gault, CKD-EPI, and MDRD) used in 8 3.1–7.6, n = 6184, 7 contributions) by the CKD-EPI for- studies. The most common methods used to assess pro- mula (Additional file 1: Figure S2, pp. 21–22). teinuria were the urine dipstick test (46 studies), followed In high-risk populations, the prevalence of CKD stages by the spot urine albumin to creatinine ratio (20 studies), 1 to 5 was 27.7% (95% CI 17.1–39.7, n = 3262, 11 contri- and the 24-h urine collection (three studies). butions) according to the MDRD equation, 49.8% (95% Kaze et al. BMC Nephrology (2018) 19:125 Page 5 of 11 Fig. 2 Prevalence of CKD in general populations of adults living on the African continent by region CI 28.8–70.8, n = 1361, six contributions) according to Table S4, pp. 15–16). However, no difference in CKD the Cockcroft-Gault formula, and 34.7% (95% CI 26.8– prevalence was seen between studies larger and smaller 43.1, n = 1234, six contributions) based on the CKD-EPI than the median sample size (p = 0.08, Additional file 1: formula (Additional file 1: Table S3, pp. 14). The preva- Table S3, pp. 14), studies with older compared to lence of CKD stages 3 to 5 in high-risk populations was younger participants (p = 0.8, Additional file 1:Table 10.6% (95% 7.6–13.9, n = 21,642, 31 contributions) ac- S3, pp. 14), or studies reported before compared to cording to the MDRD equation, 16.2% (95% CI 11.2– after the median year of publication (p = 0.33, Additional 21.9, n = 40,610, 22 contributions) according to the file 1: Table S3, pp. 14). Cockcroft-Gault formula, and 11.6% (95% CI 6.2–18.3, n = 6295, 11 contributions) based on the CKD-EPI for- Investigation of the sources of heterogeneity and mula (Additional file 1: Table S4, pp. 15–16). publication bias The prevalence of proteinuria alone was 9.8% (95% CI We found substantial heterogeneity across the contribut- 4.9–16.3, n = 8410; 11 studies) and 22.7% (95% CI 15.5– ing studies overall, within subgroups for residence, me- 30.8, n = 8784; 26 studies) in the general and high-risk dian sample size, geographic region, and across eGFR populations, respectively. estimating equations. There was no evidence of publica- tion bias across studies reporting on the prevalence of Influence of year of publication, sample size, and median CKD stages 1 to 5 in the general population, with the age Egger test for bias yielding a p-value of 0.27 (Additional The prevalence of CKD stages 3–5 was significantly file 1: Table S3, pp. 14). We found some evidence of higher in studies reported after compared to before 2013 publication bias across studies reporting on the pre- (5.7, 95% CI 4.0–7.8 vs 2.4, 95% CI 0.9–4.4, p = 0.02, valence of CKD stages 1 to 5 in high-risk populations, Additional file 1: Table S4, pp. 15–16). Likewise and as with the Egger test for bias giving a p-value of 0.01 expected, the prevalence of CKD stages 3–5 was signifi- (Additional file 1: Table S5, pp. 16–17). However, smaller cantly higher in older (≥ median age 43.7 years) com- studies (sample size < median of 192) were not more pared to younger participants (p = 0.01, Additional file 1: likely to report more extreme results compared with Kaze et al. BMC Nephrology (2018) 19:125 Page 6 of 11 Fig. 3 Prevalence of CKD in general populations of adults living in Africa. Black boxes represent the effect estimates (prevalence) and the horizontal bars are for the 95% confidence intervals (CIs). The diamond is for the pooled effect estimate and 95% CI and the dotted vertical line centered on the diamond has been added to assist visual interpretation larger studies (p-value = 0.11, Additional file 1: Table S5, limited to sub-Saharan Africa included articles published pp. 16–17). between 1962 and 2011; amongst which 32 (35.6%) were published before 2000 [9]. The vast majority (65.3%) of Discussion studies included in our review were published between Our review including 98,432 individuals found a preva- 2012 and 2016, a more contemporary period. Our find- lence of 15.8% for CKD stages 1–5 in the general popu- ings supplement previous studies and reviews on CKD lation of adults living on the African continent. by providing an updated and comprehensive synthesis of Additionally, we showed that 4.6% of adults living in data on the magnitude of CKD in the African continent. Africa have moderate or severe decreases in kidney func- The CKD prevalence in this review is slightly higher tion (i.e. CKD stages 3 to 5). The prevalence of CKD than that reported in African countries in a recently was higher in sub-Saharan Africa than North Africa, and published systematic review on the global CKD preva- nearly two times higher in high-risk populations than in lence [21]. However, their review included data from general populations. The three main equations used to three African countries (5497 individuals) and is there- estimate the kidney also yielded different results. The fore much less representative of the entire continent. Cockcroft formula showed a prevalence that was higher The overall estimated prevalence of CKD stages 1–5in than prevalence obtained using MDRD or CKD-EPI the general population found in our review is similar to equations. We found substantial heterogeneity across that (13.1%) found in the United States [22]. However, the studies and in subgroup analyses, and no evidence of the prevalence of CKD in our review was higher than publication bias across studies reporting on CKD preva- that found in general populations of adults living in four lence in general populations of Africa. Asian countries (West Malaysia, Korea, China, and Our review is the first to comprehensively assess the Taiwan) and Europe [23–27]. The prevalence of CKD prevalence of CKD in adults living on the African con- stages 3–5 in our study was lower than that found in the tinent. A previous systematic review on CKD prevalence United States (8.0%), and higher than that found in two Kaze et al. BMC Nephrology (2018) 19:125 Page 7 of 11 Fig. 4 Prevalence of CKD stages 3 to 5 in general populations of adults living in Africa. Black boxes represent the effect estimates (prevalence) and the horizontal bars are for the 95% confidence intervals (CIs). The diamond is for the pooled effect estimate and 95% CI and the dotted vertical line centered on the diamond has been added to assist visual interpretation Asian countries [23–27]. The prevalence of CKD in further compounded by the fact that a vast majority of high-risk populations found in our review is similar to people with CKD are unaware of their condition until that found in a high-risk population (with diabetes or they progress to later stages [34, 35]. Various observa- hypertension) in Korea (39.6%) [28]. Likewise, our esti- tional cohort studies have shown that the increased risk mate in high-risk subjects is comparable to the preva- in CVD mortality in CKD patients is apparent in the lence of CKD in a sample of hypertensive subjects in the early stages of the disease, and nearly 40% of deaths Unites States [29]. from CKD occur prematurely (before age 65) [26, 32]. While the attention of policy makers is finally ex- This highlights the need for interventions earlier in the tending beyond communicable diseases to the process. Effective strategies can slow the progression of non-communicable diseases, particularly the cardiovas- CKD and may help reduce the risk of CVD [14]. cular disease (CVD) epidemic, it is not fully appreciated Our review points out the critical need of data in that this is accompanied by an epidemic of CKD. Our many parts of Africa that would help to further estimates indicate that CKD may be more common than characterize the magnitude of CKD burden on the diabetes which has an estimated prevalence of 3.2% in mother continent. Indeed, out of the 54 African coun- people aged 20 to 79 in sub-Saharan Africa [30]. The tries, 32 were not included in this review. Although, the rapid rise in the number of people with hypertension or number of population-based studies on CKD prevalence diabetes [7, 8], combined with the HIV pandemics, and has somewhat increased in the recent years, many the increased survival in individuals taking antiretroviral African countries are still lagging behind. African coun- therapy are predicted to drive the burden of CKD in tries must be encouraged to conduct population-based Africa [31]. Like hypertension or diabetes, CKD has con- surveys of CKD prevalence such as the MAREMAR sistently been shown to be associated with higher risk of (Maladies Rénales Chroniques au Maroc) [36] project on mortality from CVD [26, 32, 33]. This situation is a regular basis, in order to monitor time trends of CKD Kaze et al. BMC Nephrology (2018) 19:125 Page 8 of 11 Fig. 5 Prevalence of CKD in high-risk populations of adults living in Africa. Black boxes represent the effect estimates (prevalence) and the horizontal bars are for the 95% confidence intervals (CIs). The diamond is for the pooled effect estimate and 95% CI and the dotted vertical line centered on the diamond has been added to assist visual interpretation prevalence with comparable methodologies. African not completely explained by subgroup analyses. This countries are also encouraged to incorporate CKD sur- may in part be explained by between-study differences veillance in existing data collection opportunities such in methodology and population structures, but they may as the WHO STEPwise approach to Surveillance sur- also represent true regional differences in disease bur- veys. Moreover, there is an urgent need for African na- den. Second, our ability to assess the quality of included tions to establish and sustain renal registries at both studies was limited by the incomplete methodological national and regional levels [37]. On a continent where information provided in some studies. Third, primary access to healthcare is restricted due to economic con- studies lacked data on important covariates that could straints, the publication of registry data would be a have been used in meta-regression analyses to further cost-effective approach to draw the public and policy explore and adjust for the sources of variations in preva- makers’ attention to the underappreciated problem of lence between studies. Additionally, the majority of sur- CKD, and help efforts to prevent, detect, and treat CKD veys did not follow patients for 3 months to confirm the at much earlier stages [37–39]. An African renal registry diagnosis of CKD. Previous evidence suggested that a would facilitate the sharing of expertise across all the single measurement of eGFR may overestimate CKD nations using this common platform, and lead to more prevalence [40]. These limitations notwithstanding, our effective patient advocacy, public health policy and fun- study has several strengths. First, we used a comprehen- draising [37]. sive review protocol [10], and made extensive efforts to Our review has some limitations. First, we found sub- identify all the available evidence by searching multiple stantial heterogeneity in prevalence estimates, which was electronic databases without language restrictions; we Kaze et al. BMC Nephrology (2018) 19:125 Page 9 of 11 Fig. 6 Prevalence of CKD stages 3 to 5 in high-risk populations of adults living in Africa. Black boxes represent the effect estimates (prevalence) and the horizontal bars are for the 95% confidence intervals (CIs). The diamond is for the pooled effect estimate and 95% CI and the dotted vertical line centered on the diamond has been added to assist visual interpretation applied an Africa-specific search filter [12], and adhered risks, the extent to which this applies to populations to pre-specified study selection criteria [10]. Second, we living on the African continent remains unclear. Most critically appraised the methodological quality of studies observational studies of adverse health outcomes in with a standard quality assessment tool for prevalence CKD patients were conducted in developed countries, studies [15]. Finally, we used the Freeman-Tukey single which may not necessarily generalize to African popu- arcsine transformation to stabilize the variance of pre- lations [43]. African countries are encouraged to es- valence estimates before pooling, therefore limiting the tablish prospective multicenter cohorts of CKD effects of studies with small and large prevalence esti- patients such as the CRIC (Chronic Renal Insuffi- mates on the pooled estimates [16]. ciency Cohort) Study [44], in order to examine risk Key unaddressed issues in the detection of CKD in factors for CKD progression and CVD tailored to Africa include the absence of reliable and valid methods their region, identify high-risk subgroups, and assess for assessing kidney function [41]. Our findings showed the role of genetic factors such as Apolipoprotein L1 that CKD prevalence estimates can vary substantially de- (ApoL1) [45] variants in the genesis and progression pending on the equation used. Although the MDRD and of CKD in people living on the African continent. CKD-EPI have been shown to be superior to the Cockcroft-Gault formula, the validity of those methods in African populations remains to be established [41, 42]. Conclusion Furthermore, although it is widely accepted that CKD is In summary, the substantial burden of CKD on the associated with an increased all-cause and CVD mortality African continent found in this review highlights the Kaze et al. BMC Nephrology (2018) 19:125 Page 10 of 11 need for a concerted action to prevent the high health Medical Center Midtown Campus, Baltimore, MD, USA. Division of Renal Medicine, Emory University School of Medicine, Atlanta, GA, USA. Division of and economic burden that this condition entails. Nephrology, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA. Welch Center for Additional file Prevention, Epidemiology and Clinical Research, Johns Hopkins Medical Institutions, Baltimore, MD, USA. Nephrology Center of Maryland, Baltimore, Additional file 1: Item 1: Search strategies. Item 2: Quality appraisal of MD, USA. Division of Endocrinology, Diabetes, and Hypertension, Brigham included studies. Item 3: Supplementary tables. Table S1. Summary of and Women’s Hospital, Harvard Medical School, 221 Longwood Avenue, the risk of bias in the included studies. Table S2. Characteristics of the 98 Boston, MA 02115, USA. studies included in this systematic review. Table S3. Summary statistics for prevalence of CKD stages 1–5 in general populations. Table S4. Received: 25 November 2017 Accepted: 24 May 2018 Summary statistics for prevalence of CKD stages 3–5 in general populations. Table S5. Summary statistics for prevalence of CKD stages 1–5 in high-risk populations. Table S6. Summary statistics for prevalence of CKD stages 3–5 in high-risk populations. Item 4: Supplementary figures. References Figure S1. Prevalence of CKD in the general population of Africa accord- 1. Global, regional, and national disability-adjusted life-years (DALYs) for 315 ing to eGFR equation. Figure S2. Prevalence of CKD stages 3 to 5 in the diseases and injuries and healthy life expectancy (HALE), 1990–2015. A general population of Africa according to eGFR equation. Figure S3. systematic analysis for the global burden of disease study 2015. Lancet Prevalence of CKD in HIV-positive individuals living in Africa. Figure S4. (London, England). 2016;388(10053):1603–58. Prevalence of CKD in hypertensive individuals living in Africa. Figure S5. 2. World Kidney Day: Chronic Kidney Disease. 2015; http://www. Prevalence of CKD in people with diabetes mellitus living in Africa. worldkidneyday.org/faqs/chronic-kidney-disease/ (accessed: Jan 24, 2017). Figure S6. Prevalence of CKD stages 3 to 5 in HIV-positive individuals liv- 3. Peralta CA, Risch N, Lin F, Shlipak MG, Reiner A, Ziv E, Tang H, Siscovick D, ing in Africa. Figure S7. Prevalence of CKD stages 3 to 5 in hypertensive Bibbins-Domingo K. The Association of African Ancestry and elevated individuals living in Africa. Figure S8. Prevalence of CKD stages 3 to 5 in creatinine in the coronary artery risk development in young adults (CARDIA) people with diabetes mellitus living in Africa. (PDF 1107 kb). study. Am J Nephrol. 2010;31(3):202–8. 4. Kiberd BA, Clase CM. Cumulative risk for developing end-stage renal disease Abbreviations in the US population. Journal of the American Society of Nephrology : ANOVA: Analysis of the Variance; ApoL1: Apolipoprotein L1; CKD: Chronic JASN. 2002;13(6):1635–44. kidney disease; CKD-EPI: Chronic kidney disease epidemiology collaboration; 5. Daar AS, Singer PA, Persad DL, Pramming SK, Matthews DR, Beaglehole R, CRIC: Chronic Renal Insufficiency Cohort; CVD: Cardiovascular disease; Bernstein A, Borysiewicz LK, Colagiuri S, Ganguly N, et al. Grand challenges eGFR: Estimated glomerular filtration rate; ESRD: End-stage renal disease; in chronic non-communicable diseases. Nature. 2007;450(7169):494–6. HAART: Highly active antiretroviral therapy; HIV: Human immunodeficiency 6. Naicker S. Burden of end-stage renal disease in sub-Saharan Africa. Clin virus; KDIGO: Kidney disease: improving global outcomes; Nephrol. 2010;74(Suppl 1):S13–6. MAREMAR: Maladies Rénales Chroniques au Maroc; MDRD: Modification of 7. Kaze AD, Schutte AE, Erqou S, Kengne AP, Echouffo-Tcheugui JB. Prevalence Diet in Renal Disease; NKF KDOQI: National kidney foundation Kidney disease of hypertension in older people in Africa: a systematic review and meta- outcomes quality initiative; PRISMA: Preferred Reporting Items for Systematic analysis. J Hypertens. 2017;35(7):1345–52. Reviews and Meta-analyses; RCT: Randomized controlled trial; RRT: Renal 8. Werfalli M, Engel ME, Musekiwa A, Kengne AP, Levitt NS. The prevalence of replacement therapy; WHO: World health organization type 2 diabetes among older people in Africa: a systematic review. The lancet Diabetes & endocrinology. 2016;4(1):72–84. Availability of data and materials 9. Stanifer JW, Jing B, Tolan S, Helmke N, Mukerjee R, Naicker S, Patel U. The The datasets used and analyzed during the current study are available from epidemiology of chronic kidney disease in sub-Saharan Africa: a systematic the corresponding author on reasonable request. review and meta-analysis. Lancet Glob Health. 2014;2(3):e174–81. 10. Burden of chronic kidney disease on the African continent: a systematic review and meta-analysis. https://www.crd.york.ac.uk/PROSPERO/display_ Authors’ contributions record.asp?ID=CRD42017054445 ADK contributed to protocol design, study design, the literature review, quality 11. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for assessment, data extraction, statistical analysis, data interpretation, article systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. preparation, article review, and correspondence. TI contributed to protocol design, 2009;6(7):e1000097. study design, the literature review, quality assessment, data extraction, article preparation, and article review. BGJ contributed to the data interpretation, article 12. Eisinga A, Siegfried N, Clarke M. The sensitivity and precision of search preparation, and article review. JBE contributed to protocol design, study design, terms in phases I, II and III of the Cochrane highly sensitive search strategy data interpretation, and article preparation and review. ADK, TI, BGJ and JBE take for identifying reports of randomized trials in medline in a specific area of responsibility that this study has been reported honestly, accurately, and health care–HIV/AIDS prevention and treatment interventions. Health Inf transparently; that no important aspects of the study have been omitted; and that Libr J. 2007;24(2):103–9. any discrepancies from the study as planned have been explained. All authors 13. National Kidney Foundation. K/DOQI clinical practice guidelines for chronic read and approved the final manuscript. kidney disease: evaluation, classification, and stratification. American journal of kidney diseases : the official journal of the National Kidney Foundation. 2002;39(2 Suppl 1):S1–266. Ethics approval and consent to participate 14. Stevens PE, Levin A. Evaluation and management of chronic kidney disease: Not applicable. synopsis of the kidney disease: improving global outcomes 2012 clinical practice guideline. Ann Intern Med. 2013;158(11):825–30. Competing interests 15. Hoy D, Brooks P, Woolf A, Blyth F, March L, Bain C, Baker P, Smith E, Dr. Bernard Jaar is a Section Editor of BMC Nephrology. Buchbinder R. Assessing risk of bias in prevalence studies: modification of an existing tool and evidence of interrater agreement. J Clin Epidemiol. 2012;65(9):934–9. Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in 16. Barendregt JJ, Doi SA, Lee YY, Norman RE, Vos T. Meta-analysis of published maps and institutional affiliations. prevalence. J Epidemiol Community Health. 2013;67(11):974–8. 17. Cochran WG. The combination of estimates from different experiments. Author details Biometrics. 1954;10(1):101–29. Division of Renal Medicine, Brigham and Women’s Hospital, Harvard Medical 18. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. School, Boston, MA, USA. Department of Medicine, University of Maryland Stat Med. 2002;21(11):1539–58. Kaze et al. BMC Nephrology (2018) 19:125 Page 11 of 11 19. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis 39. Davids MR, Marais N, Jacobs J. The South African renal registry: A first detected by a simple, graphical test. BMJ (Clinical research ed). 1997; report. Nephrology Dialysis Transplantation. 2014;29:iii380. 315(7109):629–34. 40. Eriksen BO, Ingebretsen OC. The progression of chronic kidney disease: a 20. Pruijm MT, Madeleine G, Riesen WF, Burnier M, Bovet P. Prevalence of 10-year population-based study of the effects of gender and age. Kidney microalbuminuria in the general population of Seychelles and strong Int. 2006;69(2):375–82. association with diabetes and hypertension independent of renal markers. J 41. Eastwood JB, Kerry SM, Plange-Rhule J, Micah FB, Antwi S, Boa FG, Banerjee Hypertens. 2008;26(5):871–7. D, Emmett L, Miller MA, Cappuccio FP. Assessment of GFR by four methods in adults in Ashanti, Ghana: the need for an eGFR equation for lean African 21. Hill NR, Fatoba ST, Oke JL, Hirst JA, O’Callaghan CA, Lasserson DS, Hobbs FD. populations. Nephrology, dialysis, transplantation : official publication of the Global prevalence of chronic kidney disease - a systematic review and European Dialysis and Transplant Association - European Renal Association. meta-analysis. PLoS One. 2016;11(7):e0158765. 2010;25(7):2178–87. 22. Coresh J, Selvin E, Stevens LA, Manzi J, Kusek JW, Eggers P, Van Lente F, 42. van Deventer HE, George JA, Paiker JE, Becker PJ, Katz IJ. Estimating Levey AS. Prevalence of chronic kidney disease in the United States. Jama. glomerular filtration rate in black south Africans by use of the modification 2007;298(17):2038–47. of diet in renal disease and Cockcroft-gault equations. Clin Chem. 2008; 23. Hooi LS, Ong LM, Ahmad G, Bavanandan S, Ahmad NA, Naidu BM, 54(7):1197–202. Mohamud WN, Yusoff MF. A population-based study measuring the 43. Matsushita K, van der Velde M, Astor BC, Woodward M, Levey AS, de Jong prevalence of chronic kidney disease among adults in West Malaysia. PE, Coresh J, Gansevoort RT. Association of estimated glomerular filtration Kidney Int. 2013;84(5):1034–40. rate and albuminuria with all-cause and cardiovascular mortality in general 24. Ji E, Kim YS. Prevalence of chronic kidney disease defined by using CKD-EPI population cohorts: a collaborative meta-analysis. Lancet (London, England). equation and albumin-to-creatinine ratio in the Korean adult population. 2010;375(9731):2073–81. The Korean journal of internal medicine. 2016;31(6):1120–30. 44. Feldman HI, Appel LJ, Chertow GM, Cifelli D, Cizman B, Daugirdas J, Fink JC, 25. Zhang L, Wang F, Wang L, Wang W, Liu B, Liu J, Chen M, He Q, Liao Y, Yu X, Franklin-Becker ED, Go AS, Hamm LL, et al. The chronic renal insufficiency et al. Prevalence of chronic kidney disease in China: a cross-sectional survey. cohort (CRIC) study: design and methods. Journal of the American Society Lancet (London, England). 2012;379(9818):815–22. of Nephrology : JASN. 2003;14(7 Suppl 2):S148–53. 26. Wen CP, Cheng TY, Tsai MK, Chang YC, Chan HT, Tsai SP, Chiang PH, Hsu 45. Dummer PD, Limou S, Rosenberg AZ, Heymann J, Nelson G, Winkler CA, CC, Sung PK, Hsu YH, et al. All-cause mortality attributable to chronic kidney Kopp JB. APOL1 kidney disease risk variants – an evolving landscape. Semin disease: a prospective cohort study based on 462 293 adults in Taiwan. Nephrol. 2015;35(3):222–36. Lancet (London, England). 2008;371(9631):2173–82. 27. Hallan SI, Coresh J, Astor BC, Asberg A, Powe NR, Romundstad S, Hallan HA, Lydersen S, Holmen J. International comparison of the relationship of chronic kidney disease prevalence and ESRD risk. Journal of the American Society of Nephrology : JASN. 2006;17(8):2275–84. 28. Kang YU, Bae EH, Ma SK, Kim SW. Determinants and burden of chronic kidney disease in a high-risk population in Korea: results from a cross- sectional study. The Korean journal of internal medicine. 2016;31(5):920–9. 29. Crews DC, Plantinga LC, Miller ER 3rd, Saran R, Hedgeman E, Saydah SH, Williams DE, Powe NR. Prevalence of chronic kidney disease in persons with undiagnosed or prehypertension in the United States. Hypertension (Dallas, Tex : 1979). 2010;55(5):1102–9. 30. IDF Diabtes Atlas - 7th Edition. http://www.diabetesatlas.org/. (accessed: Jan 12, 2017). 31. Jha V, Garcia-Garcia G, Iseki K, Li Z, Naicker S, Plattner B, Saran R, Wang AY, Yang CW. Chronic kidney disease: global dimension and perspectives. Lancet (London, England). 2013;382(9888):260–72. 32. Fried LF, Shlipak MG, Crump C, Bleyer AJ, Gottdiener JS, Kronmal RA, Kuller LH, Newman AB. Renal insufficiency as a predictor of cardiovascular outcomes and mortality in elderly individuals. J Am Coll Cardiol. 2003;41(8):1364–72. 33. Shlipak MG, Fried LF, Cushman M, Manolio TA, Peterson D, Stehman-Breen C, Bleyer A, Newman A, Siscovick D, Psaty B. Cardiovascular mortality risk in chronic kidney disease: comparison of traditional and novel risk factors. Jama. 2005;293(14):1737–45. 34. Coresh J, Byrd-Holt D, Astor BC, Briggs JP, Eggers PW, Lacher DA, Hostetter TH. Chronic kidney disease awareness, prevalence, and trends among U.S. adults, 1999 to. Journal of the American Society of Nephrology : JASN 2005. 2000;16(1):180–8. 35. Minutolo R, De Nicola L, Mazzaglia G, Postorino M, Cricelli C, Mantovani LG, Conte G, Cianciaruso B. Detection and awareness of moderate to advanced CKD by primary care practitioners: a cross-sectional study from Italy. American journal of kidney diseases : the official journal of the National Kidney Foundation. 2008;52(3):444–53. 36. Benghanem Gharbi M, Elseviers M, Zamd M, Belghiti Alaoui A, Benahadi N, Trabelssi el H, Bayahia R, Ramdani B, De Broe ME. Chronic kidney disease, hypertension, diabetes, and obesity in the adult population of Morocco: how to avoid “over”- and “under”-diagnosis of CKD. Kidney Int. 2016;89(6): 1363–71. 37. Davids MR, Eastwood JB, Selwood NH, Arogundade FA, Ashuntantang G, Benghanem Gharbi M, Jarraya F, MacPhee IA, McCulloch M, Plange-Rhule J, et al. A renal registry for Africa: first steps. Clin Kidney J. 2016;9(1):162–7. 38. Counil E, Cherni N, Kharrat M, Achour A, Trimech H. Trends of incident dialysis patients in Tunisia between 1992 and 2001. American journal of kidney diseases : the official journal of the National Kidney Foundation. 2008;51(3):463–70. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Nephrology Springer Journals

Burden of chronic kidney disease on the African continent: a systematic review and meta-analysis

Free
11 pages

Loading next page...
 
/lp/springer_journal/burden-of-chronic-kidney-disease-on-the-african-continent-a-systematic-80uSrg3wxQ
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Medicine & Public Health; Nephrology; Internal Medicine
eISSN
1471-2369
D.O.I.
10.1186/s12882-018-0930-5
Publisher site
See Article on Publisher Site

Abstract

Background: Accurate contemporary data on the burden of Chronic Kidney Disease (CKD) on the African continent are lacking. We determined the prevalence of CKD in adult populations living in Africa, and variations by stage, gender, estimated Glomerular Filtration Rate (eGFR) equation, and residence. Methods: For this systematic review, we searched multiple electronic databases for original studies on CKD prevalence reported from January 1, 2000 to December 31, 2016. Two reviewers independently undertook quality assessment and data extraction. We stabilized the variance of study-specific estimates with the Freeman-Turkey single arcsine transformation and pooled the data using a random effects meta-analysis models. Results: A total of 98 studies involving 98,432 individuals were included in the final meta-analysis. The overall prevalence was 15.8% (95% CI 12.1–19.9) for CKD stages 1–5 and 4.6% (3.3–6.1) for CKD stages 3–5 in the general population. Equivalent figures were greater at 32.3% (23.4–41.8) and 13.3% (10.7–16.0) in high-risk populations (people with hypertension, diabetes, HIV). CKD prevalence was higher in studies based on the Cockcroft-Gault formula than MDRD or CKD-EPI equations; and in studies from sub-Saharan Africa compared with those from North Africa (17.7, 95% CI 13.7–22.1 vs 6.1, 95% CI 3.6–9.3, p < 0.001). There was substantial heterogeneity across studies (all I > 90%) and no evidence of publication bias in main analyses. Conclusion: CKD is highly prevalent across Africa, inviting efforts into prevention, early detection and control of CKD in adults living on the African continent which is particularly important in a resource limited environment. Trial Registration: Prospero Registration ID: CRD42017054445. Keywords: Chronic kidney disease, Prevalence, Systematic review, Meta-analysis, Africa Background of communicable and non-communicable diseases, in Chronic kidney disease (CKD) is a leading cause of mor- part driven by the adoption of western lifestyles, changes bidity and mortality in both developed and developing in the built environment, and the rapid urbanization [5]. countries, with an estimated 10% of the population This dual burden has led to a consequential rise in the worldwide having CKD in 2015 [1, 2]. Studies have con- number of people affected by CKD on the African con- sistently shown that African descendants are at in- tinent [6]. creased risk for CKD occurrence and progression to Given the constant rise in its risk factors in Africa [7, 8], end-stage renal disease (ESRD) [3, 4]. Many African CKD is increasingly recognized as a major public health countries are currently undergoing rapid epidemiological threat, against a background of limited access to renal re- transitions and are confronted with the double burden placement therapy (RRT) [6]. Hence, in Africa, prevention and early detection of CKD in order to slow its progres- sion are of paramount importance. For this purpose, a bet- * Correspondence: jechouffotcheugui@bwh.harvard.edu ter understanding of the current prevalence of CKD in Division of Endocrinology, Diabetes, and Hypertension, Brigham and Africa is urgently needed. Although the number of reports Women’s Hospital, Harvard Medical School, 221 Longwood Avenue, Boston, MA 02115, USA on CKD prevalence across Africa has increased in recent Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Kaze et al. BMC Nephrology (2018) 19:125 Page 2 of 11 years, accurate data on its exact magnitude on the contin- according to the NKF KDOQI (National Kidney Founda- ent are still lacking [6]. The only systematic review of tion Kidney disease outcomes quality initiative or KDIGO CKD prevalence in Africa was limited to sub-Saharan (Kidney Disease: Improving Global Outcomes) guidelines countries, included studies published between 1962 and [13, 14]. We excluded studies in participants selected 2011, and highlighted the inability to make definitive in- based on the presence or absence of kidney disease, stud- ferences due to the poor quality of included studies [9]. ies limited to pregnant women. For multiple surveys con- We conducted a systematic review and meta-analysis of ducted in different countries or at different calendar years the contemporary evidence on CKD prevalence in adults and reported within the same article, each survey was living on the African continent, in order to establish base- accounted for separately, when it was possible to disaggre- line figures against which future trends can be monitored. gate the data by country. Methods Assessment of the methodological quality of included We used a prospective protocol (PROSPERO studies CRD42017054445) [10] to perform this review according Two reviewers (ADK & TI) independently assessed to the Preferred Reporting Items for Systematic Reviews study quality, with disagreements being resolved by con- and Meta-analyses (PRISMA) guidelines [11]. sensus or by consulting a third investigator (JBE). We used the ten-item rating checklist developed by Hoy et Search strategy and selection criteria al. that assesses sampling, the sampling technique and Identification of relevant studies size, outcome measurement, response rate, and statis- We performed comprehensive electronic searches of tical reporting [15]. Each item was assigned a score of 1 major databases including PubMed/Medline and Embase (yes) or 0 (no), and scores were summed across items to using an African search filter developed by Eisinga and generate an overall quality score ranging from 0 to 10. colleagues [12], to identify relevant studies published on Each study was rated as being of low, moderate, or high CKD on the African continent between January 1, 2000 methodological quality depending on the number of and December 31, 2016, without language restriction questions answered as “yes (low risk of bias)”. Studies of (Additional file 1: pp. 2–3). The searches were restricted high quality had scores higher than 8, moderate a score to the post-2000 era in order to provide the most con- of 6–8, and low a score of 5 or lower [15]. temporary estimate of CKD prevalence in Africa. Esti- mates from previous studies are also likely based on Data extraction older definitions of CKD and therefore not directly com- The same reviewers independently selected studies and parable to more recent studies. Additionally, we traced extracted relevant data using a pre-conceived extraction the citations of identified articles via the ISI Web of form. The information extracted included 1) author de- knowledge, and scanned the reference lists of review pa- tails (names and year of publication); 2) study character- pers and conference proceedings. We also searched rele- istics (country, design, setting, data source, sampling vant African journals, the World Health Organization method, sample size, data collection period, response (WHO) Global Health Library databases (which include rate); 3) Participants’ characteristics (age, gender, hyper- the African Index Medicus, WHO Library Information tension status, diabetes status, HIV status, HAART treat- System, and Scientific Electronic Library Online). ment); and 4) CKD characteristics (CKD diagnostic criteria, eGFR equation, proteinuria assessment method, Selection of included studies prevalence, number of participants tested and diagnosed Two investigators (ADK and TI) independently reviewed with CKD overall and by subgroups of interest). the articles by title, abstract, and full-text where relevant for inclusion. Disagreements were resolved by consensus Statistical analysis or by consulting a third investigator (JBE). To be included For each study, the unadjusted prevalence of CKD and in this review, primary studies had to be population or standard errors were calculated (number of cases/sample hospital- based, randomized controlled trial (RCT) or size) based on the information on crude numerators and non-randomized studies including cross-sectional, cohort, denominators provided in individual studies. CKD stages and case-cohort studies that report prevalence or enough 1–5 was defined as kidney damage (urinary dipstick ab- data to calculate prevalence of CKD in adults (aged 18 normalities) or eGFR < 60 ml/min/1.73m , while CKD and above) from sub-Saharan and North African coun- stages 3–5 was defined as eGFR < 60 ml/min/1.73m . tries. For cohort studies or RCT, we included data from When eGFR from multiple equations was reported, we the baseline evaluation. CKD had to have been defined as preferred the Chronic Kidney Disease Epidemiology the presence of kidney damage and/or estimated Glom- Collaboration (CKD-EPI) equation, then the Modifica- erular filtration rate (eGFR) < 60 mL/min/1.73 m , tion of Diet in Renal Disease (MDRD), and lastly the Kaze et al. BMC Nephrology (2018) 19:125 Page 3 of 11 Cockcroft-Gault formula in the main analyses. We used (central, eastern, northern, southern and western Africa), the DerSimonian-Laird random-effects models to gener- sample size, and year of publication. Comparisons be- ate the pooled prevalence of CKD according to each tween subgroups were performed using the Q-test based diagnostic criterion. The random effects model was on the Analysis of the Variance (ANOVA). Publication chosen in anticipation of substantial variations in CKD bias was evaluated using funnel plots supplemented by prevalence estimates across the included studies. To formal statistical assessment using the Egger’s test [19]. minimize the effect of studies with extremely small or All analyses were performed using Stata software (Stata extremely large prevalence on the overall estimate, we Corp V.14, Texas, USA). stabilized the variance of the study-specific prevalence estimates with the Freeman-Tukey single arcsine trans- Results formation before pooling the data [16]. We examined The review process prevalence by disease-specific populations (general Fig. 1 summarizes the study selection process. In total, population, HIV, Hypertension, diabetes mellitus), re- 1429 records were identified via databases searches. gion, method of kidney disease assessment, setting, date. After removing duplicates, we scanned the titles and ab- We assessed inter-rater agreement for inclusion and stracts of 1388 studies, of which 259 were selected for quality assessment using Cohen’s kappa (κ) coefficient. full-text review. Of these, 98 met the inclusion criteria We assessed heterogeneity between studies using and were retained in the final review (Fig. 1). Inter-rater Cochran’s Q statistic, H and the I statistics [17, 18], agreement for inclusion was excellent (κ = 0.965). which estimate the percentage of total variation across studies due to true between-study difference rather than Methodological quality of included studies chance, with I values of 25, 50 and 75% representing A total of 19 studies were deemed to be of high meth- low, medium and high heterogeneity, respectively. We odological quality, while 78 were categorized as being of explored sources of heterogeneity through a restriction moderate methodological quality; two studies were con- of analyses to subgroups defined by geographical area sidered to be of low quality and were therefore excluded 1429 records identified through database searches 746 from Medline 635 from Embase 48 from references and other sources 1388 records screened after duplicates removed 1129 records excluded 259 full-text articles assessed for eligibility 161 full-text articles excluded 93 no data on outcome 29 no prevalence data 22 no original research 15 duplicates 2 studies with poor quality rating 98 studies included in synthesis Fig. 1 Selection of articles for inclusion in the systematic review Included Eligibility Screening Identification Kaze et al. BMC Nephrology (2018) 19:125 Page 4 of 11 from the analysis (Additional file 1: Table S1, pp. 5–8). Prevalence of CKD in the general population Inter-rater agreement for quality assessment was excel- Fig. 2 shows the prevalence of CKD in the general popu- lent (κ = 0.910). lation by region. The overall prevalence in the general population was 15.8% (95% CI 12.1–19.9, n = 23,825, 22 studies, Fig. 3) for CKD stages 1 to 5, and 4.6% (95% CI Characteristics of included studies 3.3–6.1, n = 25,929, 27 studies, Fig. 4) for CKD stages 3 The characteristics of the included studies are summa- to 5. The overall prevalence in the general population rized in Additional file 1: Table S2 (pp 9–13). Sixty four was significantly higher in sub-Saharan Africa compared (65.3%) articles were published between 2012 and 2016. to north Africa for both of CKD stages 1–5 (17.7, 95%CI They originated from the five African sub-regions, with 13.7–22.1; 10,028 participants vs 6.1, 95% CI 3.6–9.3, the western sub-region being the most represented (37 13,797 participants, p < 0.001, Additional file 1: Table S3, studies, 37.8%) and the northern being the least repre- pp. 14–15) and CKD stages 3–5 (4.8, 95% CI 3.2–6.6, sented (six studies, 6.1%). Of the 54 African countries, n = 15,132 vs 2.6, 95% CI 2.3–2.9, n = 10,797; p =0.004, 22 (40.7%) were represented in this systematic review in- Additional file 1:Table S4, pp. 15–16). cluding Nigeria (26 studies); South Africa, Ghana (nine studies each); Tanzania (eight studies); Cameroon, Prevalence of CKD in high-risk populations Democratic Republic of Congo (six studies each); The prevalence of CKD stages 1 to 5 in the high-risk Malawi (five studies); Uganda, Zimbabwe, Egypt, Kenya populations was 32.3% (95% CI 23.4–41.8, n = 5056; 21 (three studies each); Congo, Sudan, Ethiopia, Senegal, contributions, Fig. 5) overall, 27.3% in participants with Zambia (two studies each); Burundi, Morocco, Rwanda, HIV (95% CI 17.0–38.9, n = 2007; 10 contributions, Add- Seychelles (one study each). One study presented com- itional file 1: Figure S3, pp. 23), 35.6% in participants bined data from two surveys conducted in Zimbabwe with hypertension (95% CI 27.9–43.7, n = 2199, 6 contri- and Uganda, another study presented combined data butions, Additional file 1: Figure S4, pp. 24), and 32.6% from seven sub-Saharan countries (Cameroon, Ivory (95% CI 0.3–82.3, 778 participants, four contributions, Coast Kenya, Mozambique, South Africa, Uganda, and Additional file 1: Figure S5, pp. 25) in participants with Zambia). Of the included studies, 97 were local studies, diabetes (Additional file 1: Table S5, pp. 16–17). The mainly conducted in urban settings, while only one prevalence of CKD stages 3 to 5 in the high-risk popula- study had national coverage [20]. With regards to the tions was 13.3% (95% CI 10.7–16.0, 52,353 participants; study sites, the majority were solely hospital- or 50 contributions, Fig. 6) overall, 9.1% (95% CI 6.6–11.9, clinic-based (59 studies, 60.2%), with only 39 studies be- n = 44,239, 28 contributions, Additional file 1: Figure S6, ing community-based. pp. 26) in participants with HIV, 17.9% (95% CI 10.9– Data from 98,432 participants were included, with a 26.1, n = 2971, 11 contributions, Additional file 1: Figure median age of 43 years (25th–75th percentiles: 36.6 to S7, pp. 27) in participants with hypertension, and 51.4). Fifty four (55.1%) studies examined participants 22.0% (95% CI 16.1–28.6, n = 5071, 10 contributions, thought be at high risk of CKD (people with hyperten- Additional file 1: Figure S8, pp. 28) in participants sion, diabetes, HIV, or sickle cell disease), while 33 with diabetes (Additional file 1: Table S6, pp. 17–18). (33.7%) were conducted in the general population or subjects not known to be at risk of CKD, and 11 (11.2%) Prevalence of CKD by eGFR estimating equation studies in both. The included studies applied various es- In general populations, the prevalence of CKD stages 1–5 timators of GFR. Fifty eight applied a single equation; was 13.7% (95% CI 10.2–17.6, n = 23,825, 14 contribu- the most frequently used being the Modification of Diet tions) according to the MDRD equation, 21.3% (95% CI in Renal Disease (MDRD) eq. (32 studies), followed by 9.9–35.5 n = 5316, 8 contributions) according to the the Cockcroft-Gault formula (17 studies), the Chronic Cockcroft-Gault formula, and 19.5% (95% CI 13.8–25.9, Kidney Disease Epidemiology Collaboration (CKD-EPI) n = 4854, 5 contributions) when using the CKD-EPI formula (7 studies), and the Cystatin C equation (two formula (Additional file 1:FigureS1, pp. 19–20). The studies). Of the 22 studies that compared eGFR using prevalence of CKD stages 3–5 was 4.5% (95% CI 3.0–6.3, two or more equations, Cockcroft-Gault and MDRD n = 22,969, 21 contributions) by the MDRD equation, equations were used in 11 studies, CKD-EPI and 11.8% (95% CI 7.3–17.1, n = 4927, 11 contributions) ac- MDRD used in 3 studies, and the three formulas cording to the Cockcroft-Gault formula, and 5.1% (95% CI (Cockcroft-Gault, CKD-EPI, and MDRD) used in 8 3.1–7.6, n = 6184, 7 contributions) by the CKD-EPI for- studies. The most common methods used to assess pro- mula (Additional file 1: Figure S2, pp. 21–22). teinuria were the urine dipstick test (46 studies), followed In high-risk populations, the prevalence of CKD stages by the spot urine albumin to creatinine ratio (20 studies), 1 to 5 was 27.7% (95% CI 17.1–39.7, n = 3262, 11 contri- and the 24-h urine collection (three studies). butions) according to the MDRD equation, 49.8% (95% Kaze et al. BMC Nephrology (2018) 19:125 Page 5 of 11 Fig. 2 Prevalence of CKD in general populations of adults living on the African continent by region CI 28.8–70.8, n = 1361, six contributions) according to Table S4, pp. 15–16). However, no difference in CKD the Cockcroft-Gault formula, and 34.7% (95% CI 26.8– prevalence was seen between studies larger and smaller 43.1, n = 1234, six contributions) based on the CKD-EPI than the median sample size (p = 0.08, Additional file 1: formula (Additional file 1: Table S3, pp. 14). The preva- Table S3, pp. 14), studies with older compared to lence of CKD stages 3 to 5 in high-risk populations was younger participants (p = 0.8, Additional file 1:Table 10.6% (95% 7.6–13.9, n = 21,642, 31 contributions) ac- S3, pp. 14), or studies reported before compared to cording to the MDRD equation, 16.2% (95% CI 11.2– after the median year of publication (p = 0.33, Additional 21.9, n = 40,610, 22 contributions) according to the file 1: Table S3, pp. 14). Cockcroft-Gault formula, and 11.6% (95% CI 6.2–18.3, n = 6295, 11 contributions) based on the CKD-EPI for- Investigation of the sources of heterogeneity and mula (Additional file 1: Table S4, pp. 15–16). publication bias The prevalence of proteinuria alone was 9.8% (95% CI We found substantial heterogeneity across the contribut- 4.9–16.3, n = 8410; 11 studies) and 22.7% (95% CI 15.5– ing studies overall, within subgroups for residence, me- 30.8, n = 8784; 26 studies) in the general and high-risk dian sample size, geographic region, and across eGFR populations, respectively. estimating equations. There was no evidence of publica- tion bias across studies reporting on the prevalence of Influence of year of publication, sample size, and median CKD stages 1 to 5 in the general population, with the age Egger test for bias yielding a p-value of 0.27 (Additional The prevalence of CKD stages 3–5 was significantly file 1: Table S3, pp. 14). We found some evidence of higher in studies reported after compared to before 2013 publication bias across studies reporting on the pre- (5.7, 95% CI 4.0–7.8 vs 2.4, 95% CI 0.9–4.4, p = 0.02, valence of CKD stages 1 to 5 in high-risk populations, Additional file 1: Table S4, pp. 15–16). Likewise and as with the Egger test for bias giving a p-value of 0.01 expected, the prevalence of CKD stages 3–5 was signifi- (Additional file 1: Table S5, pp. 16–17). However, smaller cantly higher in older (≥ median age 43.7 years) com- studies (sample size < median of 192) were not more pared to younger participants (p = 0.01, Additional file 1: likely to report more extreme results compared with Kaze et al. BMC Nephrology (2018) 19:125 Page 6 of 11 Fig. 3 Prevalence of CKD in general populations of adults living in Africa. Black boxes represent the effect estimates (prevalence) and the horizontal bars are for the 95% confidence intervals (CIs). The diamond is for the pooled effect estimate and 95% CI and the dotted vertical line centered on the diamond has been added to assist visual interpretation larger studies (p-value = 0.11, Additional file 1: Table S5, limited to sub-Saharan Africa included articles published pp. 16–17). between 1962 and 2011; amongst which 32 (35.6%) were published before 2000 [9]. The vast majority (65.3%) of Discussion studies included in our review were published between Our review including 98,432 individuals found a preva- 2012 and 2016, a more contemporary period. Our find- lence of 15.8% for CKD stages 1–5 in the general popu- ings supplement previous studies and reviews on CKD lation of adults living on the African continent. by providing an updated and comprehensive synthesis of Additionally, we showed that 4.6% of adults living in data on the magnitude of CKD in the African continent. Africa have moderate or severe decreases in kidney func- The CKD prevalence in this review is slightly higher tion (i.e. CKD stages 3 to 5). The prevalence of CKD than that reported in African countries in a recently was higher in sub-Saharan Africa than North Africa, and published systematic review on the global CKD preva- nearly two times higher in high-risk populations than in lence [21]. However, their review included data from general populations. The three main equations used to three African countries (5497 individuals) and is there- estimate the kidney also yielded different results. The fore much less representative of the entire continent. Cockcroft formula showed a prevalence that was higher The overall estimated prevalence of CKD stages 1–5in than prevalence obtained using MDRD or CKD-EPI the general population found in our review is similar to equations. We found substantial heterogeneity across that (13.1%) found in the United States [22]. However, the studies and in subgroup analyses, and no evidence of the prevalence of CKD in our review was higher than publication bias across studies reporting on CKD preva- that found in general populations of adults living in four lence in general populations of Africa. Asian countries (West Malaysia, Korea, China, and Our review is the first to comprehensively assess the Taiwan) and Europe [23–27]. The prevalence of CKD prevalence of CKD in adults living on the African con- stages 3–5 in our study was lower than that found in the tinent. A previous systematic review on CKD prevalence United States (8.0%), and higher than that found in two Kaze et al. BMC Nephrology (2018) 19:125 Page 7 of 11 Fig. 4 Prevalence of CKD stages 3 to 5 in general populations of adults living in Africa. Black boxes represent the effect estimates (prevalence) and the horizontal bars are for the 95% confidence intervals (CIs). The diamond is for the pooled effect estimate and 95% CI and the dotted vertical line centered on the diamond has been added to assist visual interpretation Asian countries [23–27]. The prevalence of CKD in further compounded by the fact that a vast majority of high-risk populations found in our review is similar to people with CKD are unaware of their condition until that found in a high-risk population (with diabetes or they progress to later stages [34, 35]. Various observa- hypertension) in Korea (39.6%) [28]. Likewise, our esti- tional cohort studies have shown that the increased risk mate in high-risk subjects is comparable to the preva- in CVD mortality in CKD patients is apparent in the lence of CKD in a sample of hypertensive subjects in the early stages of the disease, and nearly 40% of deaths Unites States [29]. from CKD occur prematurely (before age 65) [26, 32]. While the attention of policy makers is finally ex- This highlights the need for interventions earlier in the tending beyond communicable diseases to the process. Effective strategies can slow the progression of non-communicable diseases, particularly the cardiovas- CKD and may help reduce the risk of CVD [14]. cular disease (CVD) epidemic, it is not fully appreciated Our review points out the critical need of data in that this is accompanied by an epidemic of CKD. Our many parts of Africa that would help to further estimates indicate that CKD may be more common than characterize the magnitude of CKD burden on the diabetes which has an estimated prevalence of 3.2% in mother continent. Indeed, out of the 54 African coun- people aged 20 to 79 in sub-Saharan Africa [30]. The tries, 32 were not included in this review. Although, the rapid rise in the number of people with hypertension or number of population-based studies on CKD prevalence diabetes [7, 8], combined with the HIV pandemics, and has somewhat increased in the recent years, many the increased survival in individuals taking antiretroviral African countries are still lagging behind. African coun- therapy are predicted to drive the burden of CKD in tries must be encouraged to conduct population-based Africa [31]. Like hypertension or diabetes, CKD has con- surveys of CKD prevalence such as the MAREMAR sistently been shown to be associated with higher risk of (Maladies Rénales Chroniques au Maroc) [36] project on mortality from CVD [26, 32, 33]. This situation is a regular basis, in order to monitor time trends of CKD Kaze et al. BMC Nephrology (2018) 19:125 Page 8 of 11 Fig. 5 Prevalence of CKD in high-risk populations of adults living in Africa. Black boxes represent the effect estimates (prevalence) and the horizontal bars are for the 95% confidence intervals (CIs). The diamond is for the pooled effect estimate and 95% CI and the dotted vertical line centered on the diamond has been added to assist visual interpretation prevalence with comparable methodologies. African not completely explained by subgroup analyses. This countries are also encouraged to incorporate CKD sur- may in part be explained by between-study differences veillance in existing data collection opportunities such in methodology and population structures, but they may as the WHO STEPwise approach to Surveillance sur- also represent true regional differences in disease bur- veys. Moreover, there is an urgent need for African na- den. Second, our ability to assess the quality of included tions to establish and sustain renal registries at both studies was limited by the incomplete methodological national and regional levels [37]. On a continent where information provided in some studies. Third, primary access to healthcare is restricted due to economic con- studies lacked data on important covariates that could straints, the publication of registry data would be a have been used in meta-regression analyses to further cost-effective approach to draw the public and policy explore and adjust for the sources of variations in preva- makers’ attention to the underappreciated problem of lence between studies. Additionally, the majority of sur- CKD, and help efforts to prevent, detect, and treat CKD veys did not follow patients for 3 months to confirm the at much earlier stages [37–39]. An African renal registry diagnosis of CKD. Previous evidence suggested that a would facilitate the sharing of expertise across all the single measurement of eGFR may overestimate CKD nations using this common platform, and lead to more prevalence [40]. These limitations notwithstanding, our effective patient advocacy, public health policy and fun- study has several strengths. First, we used a comprehen- draising [37]. sive review protocol [10], and made extensive efforts to Our review has some limitations. First, we found sub- identify all the available evidence by searching multiple stantial heterogeneity in prevalence estimates, which was electronic databases without language restrictions; we Kaze et al. BMC Nephrology (2018) 19:125 Page 9 of 11 Fig. 6 Prevalence of CKD stages 3 to 5 in high-risk populations of adults living in Africa. Black boxes represent the effect estimates (prevalence) and the horizontal bars are for the 95% confidence intervals (CIs). The diamond is for the pooled effect estimate and 95% CI and the dotted vertical line centered on the diamond has been added to assist visual interpretation applied an Africa-specific search filter [12], and adhered risks, the extent to which this applies to populations to pre-specified study selection criteria [10]. Second, we living on the African continent remains unclear. Most critically appraised the methodological quality of studies observational studies of adverse health outcomes in with a standard quality assessment tool for prevalence CKD patients were conducted in developed countries, studies [15]. Finally, we used the Freeman-Tukey single which may not necessarily generalize to African popu- arcsine transformation to stabilize the variance of pre- lations [43]. African countries are encouraged to es- valence estimates before pooling, therefore limiting the tablish prospective multicenter cohorts of CKD effects of studies with small and large prevalence esti- patients such as the CRIC (Chronic Renal Insuffi- mates on the pooled estimates [16]. ciency Cohort) Study [44], in order to examine risk Key unaddressed issues in the detection of CKD in factors for CKD progression and CVD tailored to Africa include the absence of reliable and valid methods their region, identify high-risk subgroups, and assess for assessing kidney function [41]. Our findings showed the role of genetic factors such as Apolipoprotein L1 that CKD prevalence estimates can vary substantially de- (ApoL1) [45] variants in the genesis and progression pending on the equation used. Although the MDRD and of CKD in people living on the African continent. CKD-EPI have been shown to be superior to the Cockcroft-Gault formula, the validity of those methods in African populations remains to be established [41, 42]. Conclusion Furthermore, although it is widely accepted that CKD is In summary, the substantial burden of CKD on the associated with an increased all-cause and CVD mortality African continent found in this review highlights the Kaze et al. BMC Nephrology (2018) 19:125 Page 10 of 11 need for a concerted action to prevent the high health Medical Center Midtown Campus, Baltimore, MD, USA. Division of Renal Medicine, Emory University School of Medicine, Atlanta, GA, USA. Division of and economic burden that this condition entails. Nephrology, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA. Welch Center for Additional file Prevention, Epidemiology and Clinical Research, Johns Hopkins Medical Institutions, Baltimore, MD, USA. Nephrology Center of Maryland, Baltimore, Additional file 1: Item 1: Search strategies. Item 2: Quality appraisal of MD, USA. Division of Endocrinology, Diabetes, and Hypertension, Brigham included studies. Item 3: Supplementary tables. Table S1. Summary of and Women’s Hospital, Harvard Medical School, 221 Longwood Avenue, the risk of bias in the included studies. Table S2. Characteristics of the 98 Boston, MA 02115, USA. studies included in this systematic review. Table S3. Summary statistics for prevalence of CKD stages 1–5 in general populations. Table S4. Received: 25 November 2017 Accepted: 24 May 2018 Summary statistics for prevalence of CKD stages 3–5 in general populations. Table S5. Summary statistics for prevalence of CKD stages 1–5 in high-risk populations. Table S6. Summary statistics for prevalence of CKD stages 3–5 in high-risk populations. Item 4: Supplementary figures. References Figure S1. Prevalence of CKD in the general population of Africa accord- 1. Global, regional, and national disability-adjusted life-years (DALYs) for 315 ing to eGFR equation. Figure S2. Prevalence of CKD stages 3 to 5 in the diseases and injuries and healthy life expectancy (HALE), 1990–2015. A general population of Africa according to eGFR equation. Figure S3. systematic analysis for the global burden of disease study 2015. Lancet Prevalence of CKD in HIV-positive individuals living in Africa. Figure S4. (London, England). 2016;388(10053):1603–58. Prevalence of CKD in hypertensive individuals living in Africa. Figure S5. 2. World Kidney Day: Chronic Kidney Disease. 2015; http://www. Prevalence of CKD in people with diabetes mellitus living in Africa. worldkidneyday.org/faqs/chronic-kidney-disease/ (accessed: Jan 24, 2017). Figure S6. Prevalence of CKD stages 3 to 5 in HIV-positive individuals liv- 3. Peralta CA, Risch N, Lin F, Shlipak MG, Reiner A, Ziv E, Tang H, Siscovick D, ing in Africa. Figure S7. Prevalence of CKD stages 3 to 5 in hypertensive Bibbins-Domingo K. The Association of African Ancestry and elevated individuals living in Africa. Figure S8. Prevalence of CKD stages 3 to 5 in creatinine in the coronary artery risk development in young adults (CARDIA) people with diabetes mellitus living in Africa. (PDF 1107 kb). study. Am J Nephrol. 2010;31(3):202–8. 4. Kiberd BA, Clase CM. Cumulative risk for developing end-stage renal disease Abbreviations in the US population. Journal of the American Society of Nephrology : ANOVA: Analysis of the Variance; ApoL1: Apolipoprotein L1; CKD: Chronic JASN. 2002;13(6):1635–44. kidney disease; CKD-EPI: Chronic kidney disease epidemiology collaboration; 5. Daar AS, Singer PA, Persad DL, Pramming SK, Matthews DR, Beaglehole R, CRIC: Chronic Renal Insufficiency Cohort; CVD: Cardiovascular disease; Bernstein A, Borysiewicz LK, Colagiuri S, Ganguly N, et al. Grand challenges eGFR: Estimated glomerular filtration rate; ESRD: End-stage renal disease; in chronic non-communicable diseases. Nature. 2007;450(7169):494–6. HAART: Highly active antiretroviral therapy; HIV: Human immunodeficiency 6. Naicker S. Burden of end-stage renal disease in sub-Saharan Africa. Clin virus; KDIGO: Kidney disease: improving global outcomes; Nephrol. 2010;74(Suppl 1):S13–6. MAREMAR: Maladies Rénales Chroniques au Maroc; MDRD: Modification of 7. Kaze AD, Schutte AE, Erqou S, Kengne AP, Echouffo-Tcheugui JB. Prevalence Diet in Renal Disease; NKF KDOQI: National kidney foundation Kidney disease of hypertension in older people in Africa: a systematic review and meta- outcomes quality initiative; PRISMA: Preferred Reporting Items for Systematic analysis. J Hypertens. 2017;35(7):1345–52. Reviews and Meta-analyses; RCT: Randomized controlled trial; RRT: Renal 8. Werfalli M, Engel ME, Musekiwa A, Kengne AP, Levitt NS. The prevalence of replacement therapy; WHO: World health organization type 2 diabetes among older people in Africa: a systematic review. The lancet Diabetes & endocrinology. 2016;4(1):72–84. Availability of data and materials 9. Stanifer JW, Jing B, Tolan S, Helmke N, Mukerjee R, Naicker S, Patel U. The The datasets used and analyzed during the current study are available from epidemiology of chronic kidney disease in sub-Saharan Africa: a systematic the corresponding author on reasonable request. review and meta-analysis. Lancet Glob Health. 2014;2(3):e174–81. 10. Burden of chronic kidney disease on the African continent: a systematic review and meta-analysis. https://www.crd.york.ac.uk/PROSPERO/display_ Authors’ contributions record.asp?ID=CRD42017054445 ADK contributed to protocol design, study design, the literature review, quality 11. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for assessment, data extraction, statistical analysis, data interpretation, article systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. preparation, article review, and correspondence. TI contributed to protocol design, 2009;6(7):e1000097. study design, the literature review, quality assessment, data extraction, article preparation, and article review. BGJ contributed to the data interpretation, article 12. Eisinga A, Siegfried N, Clarke M. The sensitivity and precision of search preparation, and article review. JBE contributed to protocol design, study design, terms in phases I, II and III of the Cochrane highly sensitive search strategy data interpretation, and article preparation and review. ADK, TI, BGJ and JBE take for identifying reports of randomized trials in medline in a specific area of responsibility that this study has been reported honestly, accurately, and health care–HIV/AIDS prevention and treatment interventions. Health Inf transparently; that no important aspects of the study have been omitted; and that Libr J. 2007;24(2):103–9. any discrepancies from the study as planned have been explained. All authors 13. National Kidney Foundation. K/DOQI clinical practice guidelines for chronic read and approved the final manuscript. kidney disease: evaluation, classification, and stratification. American journal of kidney diseases : the official journal of the National Kidney Foundation. 2002;39(2 Suppl 1):S1–266. Ethics approval and consent to participate 14. Stevens PE, Levin A. Evaluation and management of chronic kidney disease: Not applicable. synopsis of the kidney disease: improving global outcomes 2012 clinical practice guideline. Ann Intern Med. 2013;158(11):825–30. Competing interests 15. Hoy D, Brooks P, Woolf A, Blyth F, March L, Bain C, Baker P, Smith E, Dr. Bernard Jaar is a Section Editor of BMC Nephrology. Buchbinder R. Assessing risk of bias in prevalence studies: modification of an existing tool and evidence of interrater agreement. J Clin Epidemiol. 2012;65(9):934–9. Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in 16. Barendregt JJ, Doi SA, Lee YY, Norman RE, Vos T. Meta-analysis of published maps and institutional affiliations. prevalence. J Epidemiol Community Health. 2013;67(11):974–8. 17. Cochran WG. The combination of estimates from different experiments. Author details Biometrics. 1954;10(1):101–29. Division of Renal Medicine, Brigham and Women’s Hospital, Harvard Medical 18. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. School, Boston, MA, USA. Department of Medicine, University of Maryland Stat Med. 2002;21(11):1539–58. Kaze et al. BMC Nephrology (2018) 19:125 Page 11 of 11 19. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis 39. Davids MR, Marais N, Jacobs J. The South African renal registry: A first detected by a simple, graphical test. BMJ (Clinical research ed). 1997; report. Nephrology Dialysis Transplantation. 2014;29:iii380. 315(7109):629–34. 40. Eriksen BO, Ingebretsen OC. The progression of chronic kidney disease: a 20. Pruijm MT, Madeleine G, Riesen WF, Burnier M, Bovet P. Prevalence of 10-year population-based study of the effects of gender and age. Kidney microalbuminuria in the general population of Seychelles and strong Int. 2006;69(2):375–82. association with diabetes and hypertension independent of renal markers. J 41. Eastwood JB, Kerry SM, Plange-Rhule J, Micah FB, Antwi S, Boa FG, Banerjee Hypertens. 2008;26(5):871–7. D, Emmett L, Miller MA, Cappuccio FP. Assessment of GFR by four methods in adults in Ashanti, Ghana: the need for an eGFR equation for lean African 21. Hill NR, Fatoba ST, Oke JL, Hirst JA, O’Callaghan CA, Lasserson DS, Hobbs FD. populations. Nephrology, dialysis, transplantation : official publication of the Global prevalence of chronic kidney disease - a systematic review and European Dialysis and Transplant Association - European Renal Association. meta-analysis. PLoS One. 2016;11(7):e0158765. 2010;25(7):2178–87. 22. Coresh J, Selvin E, Stevens LA, Manzi J, Kusek JW, Eggers P, Van Lente F, 42. van Deventer HE, George JA, Paiker JE, Becker PJ, Katz IJ. Estimating Levey AS. Prevalence of chronic kidney disease in the United States. Jama. glomerular filtration rate in black south Africans by use of the modification 2007;298(17):2038–47. of diet in renal disease and Cockcroft-gault equations. Clin Chem. 2008; 23. Hooi LS, Ong LM, Ahmad G, Bavanandan S, Ahmad NA, Naidu BM, 54(7):1197–202. Mohamud WN, Yusoff MF. A population-based study measuring the 43. Matsushita K, van der Velde M, Astor BC, Woodward M, Levey AS, de Jong prevalence of chronic kidney disease among adults in West Malaysia. PE, Coresh J, Gansevoort RT. Association of estimated glomerular filtration Kidney Int. 2013;84(5):1034–40. rate and albuminuria with all-cause and cardiovascular mortality in general 24. Ji E, Kim YS. Prevalence of chronic kidney disease defined by using CKD-EPI population cohorts: a collaborative meta-analysis. Lancet (London, England). equation and albumin-to-creatinine ratio in the Korean adult population. 2010;375(9731):2073–81. The Korean journal of internal medicine. 2016;31(6):1120–30. 44. Feldman HI, Appel LJ, Chertow GM, Cifelli D, Cizman B, Daugirdas J, Fink JC, 25. Zhang L, Wang F, Wang L, Wang W, Liu B, Liu J, Chen M, He Q, Liao Y, Yu X, Franklin-Becker ED, Go AS, Hamm LL, et al. The chronic renal insufficiency et al. Prevalence of chronic kidney disease in China: a cross-sectional survey. cohort (CRIC) study: design and methods. Journal of the American Society Lancet (London, England). 2012;379(9818):815–22. of Nephrology : JASN. 2003;14(7 Suppl 2):S148–53. 26. Wen CP, Cheng TY, Tsai MK, Chang YC, Chan HT, Tsai SP, Chiang PH, Hsu 45. Dummer PD, Limou S, Rosenberg AZ, Heymann J, Nelson G, Winkler CA, CC, Sung PK, Hsu YH, et al. All-cause mortality attributable to chronic kidney Kopp JB. APOL1 kidney disease risk variants – an evolving landscape. Semin disease: a prospective cohort study based on 462 293 adults in Taiwan. Nephrol. 2015;35(3):222–36. Lancet (London, England). 2008;371(9631):2173–82. 27. Hallan SI, Coresh J, Astor BC, Asberg A, Powe NR, Romundstad S, Hallan HA, Lydersen S, Holmen J. International comparison of the relationship of chronic kidney disease prevalence and ESRD risk. Journal of the American Society of Nephrology : JASN. 2006;17(8):2275–84. 28. Kang YU, Bae EH, Ma SK, Kim SW. Determinants and burden of chronic kidney disease in a high-risk population in Korea: results from a cross- sectional study. The Korean journal of internal medicine. 2016;31(5):920–9. 29. Crews DC, Plantinga LC, Miller ER 3rd, Saran R, Hedgeman E, Saydah SH, Williams DE, Powe NR. Prevalence of chronic kidney disease in persons with undiagnosed or prehypertension in the United States. Hypertension (Dallas, Tex : 1979). 2010;55(5):1102–9. 30. IDF Diabtes Atlas - 7th Edition. http://www.diabetesatlas.org/. (accessed: Jan 12, 2017). 31. Jha V, Garcia-Garcia G, Iseki K, Li Z, Naicker S, Plattner B, Saran R, Wang AY, Yang CW. Chronic kidney disease: global dimension and perspectives. Lancet (London, England). 2013;382(9888):260–72. 32. Fried LF, Shlipak MG, Crump C, Bleyer AJ, Gottdiener JS, Kronmal RA, Kuller LH, Newman AB. Renal insufficiency as a predictor of cardiovascular outcomes and mortality in elderly individuals. J Am Coll Cardiol. 2003;41(8):1364–72. 33. Shlipak MG, Fried LF, Cushman M, Manolio TA, Peterson D, Stehman-Breen C, Bleyer A, Newman A, Siscovick D, Psaty B. Cardiovascular mortality risk in chronic kidney disease: comparison of traditional and novel risk factors. Jama. 2005;293(14):1737–45. 34. Coresh J, Byrd-Holt D, Astor BC, Briggs JP, Eggers PW, Lacher DA, Hostetter TH. Chronic kidney disease awareness, prevalence, and trends among U.S. adults, 1999 to. Journal of the American Society of Nephrology : JASN 2005. 2000;16(1):180–8. 35. Minutolo R, De Nicola L, Mazzaglia G, Postorino M, Cricelli C, Mantovani LG, Conte G, Cianciaruso B. Detection and awareness of moderate to advanced CKD by primary care practitioners: a cross-sectional study from Italy. American journal of kidney diseases : the official journal of the National Kidney Foundation. 2008;52(3):444–53. 36. Benghanem Gharbi M, Elseviers M, Zamd M, Belghiti Alaoui A, Benahadi N, Trabelssi el H, Bayahia R, Ramdani B, De Broe ME. Chronic kidney disease, hypertension, diabetes, and obesity in the adult population of Morocco: how to avoid “over”- and “under”-diagnosis of CKD. Kidney Int. 2016;89(6): 1363–71. 37. Davids MR, Eastwood JB, Selwood NH, Arogundade FA, Ashuntantang G, Benghanem Gharbi M, Jarraya F, MacPhee IA, McCulloch M, Plange-Rhule J, et al. A renal registry for Africa: first steps. Clin Kidney J. 2016;9(1):162–7. 38. Counil E, Cherni N, Kharrat M, Achour A, Trimech H. Trends of incident dialysis patients in Tunisia between 1992 and 2001. American journal of kidney diseases : the official journal of the National Kidney Foundation. 2008;51(3):463–70.

Journal

BMC NephrologySpringer Journals

Published: Jun 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off