Prevalence of Microalbuminuria Among Diabetes Patients in Africa: A Systematic Review and Meta-Analysis

Abstract

Background

Microalbuminuria (MAU) is considered the earliest sign of diabetic nephropathy among diabetes patients. In order to effectively manage diabetic nephropathy and its consequences early, detection of microalbuminuria as soon as possible, especially for diabetes patients, is critical. Therefore, the present study aimed to determine the pooled prevalence of microalbuminuria among diabetes patients in Africa.

Methods

Electronic databases such as Google Scholar, PubMed, African Journals Online, Web of Science, Cochrane Library, EMBASE, and ResearchGate were searched for articles and grey literature. The STATA version 14 software was used to conduct the meta-analysis. I² and Cochran’s Q test were employed to assess the presence of heterogeneity between studies. Due to the presence of heterogeneity, a random effect model was used. The publication bias was assessed using the symmetry of the funnel plot and Egger’s test statistics. Moreover, subgroup analysis, trim and fill analysis, and sensitivity analysis were also done.

Results

The overall pooled prevalence of microalbuminuria among diabetes patients in Africa was 37.11% (95% CI 31.27–42.95). Substantial heterogeneity was observed between studies, with I² values of 94.7%. Moreover, this meta-analysis showed that the pooled estimate of microalbuminuria among type 1 and type 2 diabetes patients was 35.34% (95% CI: 23.89–46.80, I²=94.2), and 40.24% (95% CI: 32.0–48.47, I²=94.9) respectively. MAU, on the other hand, was more common in people with diabetes for more than 5 years 38.73% (95% CI: 29.34–48.13) than in people with diabetes for less than 5 years 31.48% (95% CI: 18.73–44.23).

Conclusion

This systematic review and meta-analysis found a high prevalence of microalbuminuria among diabetes patients. As a result, early detection of microalbuminuria is critical for preventing and treating microvascular complications such as diabetic nephropathy and the onset of end-stage renal disease.

Keywords:

• microalbuminuria
• diabetes
• Africa
• meta-analysis

Introduction

Globally, the prevalence of diabetes is increasing at an alarming rate and has become a public health concern.^1–3 According to the 10th edition of the International Diabetes Federation (IDF) Diabetes Atlas, there will be 537 million people living with diabetes worldwide in 2021. The global prevalence of diabetes is now estimated to be over 10%.⁴ In Africa, the number of people with diabetes is expected to increase by 162.5% by the year 2045.⁵ Diabetes mellitus (DM) is associated with the derangement of the normal metabolism of carbohydrates, lipids, and proteins. The primary cause of morbidity and mortality in DM is macrovascular and microvascular complications.⁶ Long-term complications of diabetes can lead to visual impairment (retinopathy), blindness, kidney disease (nephropathy), nerve damage, amputation, heart disease, and stroke.⁷ Diabetic nephropathy (DN) is a common complication in diabetic patients characterised by persistent albuminuria, a progressive decline in glomerular filtration rate (GFR), and raised arterial blood pressure.⁸ Diabetic nephropathy is the leading cause of end-stage renal disease and premature mortality in diabetic patients due to its insidious onset.^9–13 Approximately one-third to half of patients with diabetes develop renal manifestations.^14,15

According to studies, 20 to 40% of type 2 DM patients eventually develop nephropathy.^16,17 The development of DN consists of several stages, the earliest being microalbuminuria, which can progress to overt proteinuria and ultimately end-stage renal disease (ESRD).^18,19 Microalbuminuria (MAU) remains the best-documented predictor of the high risk of the development of diabetic nephropathy in DM patients.²⁰ It can be defined as a urinary albumin excretion rate (UAE) of between 30–300 mg/24 hours or as an albumin/creatinine ratio (ACR) of 30–300 mg albumin/g of creatinine.²¹ According to the theory that MAU does reflect both phases of glomerular failure and generalised endothelial dysfunction, MAU and extra renal vascular damage in diabetic patients are related.^22–24

MAU in diabetic patients is more likely to progress to overt proteinuria and, eventually, renal failure.^25–28 Typically, diabetic nephropathy advances irreversibly from the onset of clinical proteinuria to ESRD. However, it has been demonstrated that early detection, medical care, and appropriate lifestyle changes can stop or reverse the progression from micro- to macroalbuminuria.²⁹ According to some data, after 10 to 15 years of untreated type 1 diabetes with persistent MAU, over 80% of patients will have overt nephropathy, and 50% will eventually progress to end-stage renal disease (ESRD).³⁰ 20–40% of type 2 diabetes individuals with MAU advance to overt nephropathy after 20 years from the time of onset, and about 20% develop ESRD, according to research.³¹

Renal replacement therapy is not widely available in African settings. Therefore, it is essential to identify microalbuminuria as soon as possible, especially in this high-risk group of people.³² Therefore, MAU is a highly valuable indicator of kidney health and therapy outcomes in diabetic patients. There have been a lot of studies published that evaluate the prevalence of MAU in African DM patients. However, the majority of these studies used a limited sample size and only one healthcare facility, and they revealed a widely varied frequency of MAU. As a result, the current study was conducted with the goal of merging studies from the existing literature to evaluate the combined prevalence of microalbuminuria in African diabetic patients.

Methods

Search Strategy and Selection Criteria

This systematic review and meta-analysis were registered at PROSPERO with registration ID 2022: CRD42022344430. In accordance with the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines, the current review was conducted.³³ Systematic electronic searches using databases such as Google Scholar, PubMed, African Journals Online, Web of Science, Cochrane Library, EMBASE, and ResearchGate were done from May to July 4, 2022, to retrieve all relevant primary articles reporting the prevalence of MAU among diabetes patients in Africa. In addition, we extended our search by retrieving reference lists of eligible studies. The search protocol was formulated using the following keywords combined by Boolean logic (AND/OR): microalbuminuria, diabetes, diabetes mellitus, type 1 diabetes, non-insulin-dependent diabetes mellitus, type 2 diabetes, insulin-dependent diabetes mellitus, and each African country: Searched articles were entered into Endnote Software to avoid duplicates, and the list was consolidated into one. The two reviewers (OM and EA) blindly screened the titles, abstracts, and full-text search results to identify potentially eligible studies. And the full text of selected articles was assessed in detail against the inclusion criteria.

Eligibility Criteria

Inclusion Criteria

Full-length articles that report the prevalence of microalbuminuria and/or are able to calculate the prevalence of MAU among DM patients were included. Studies that reported increased urinary albumin excretion using a urinary albumin excretion rate (UAE) of between 30–299 mg/24 hours or an albumin/creatinine ratio (ACR) of 30–299 mg albumin/g of creatinine were included. Furthermore, grey literature written in the English language was also included.

Exclusion Criteria

Articles whose full length were not available and lacked the necessary data to be extracted, as well as full-length articles published in a language other than English, were excluded. Case reports, clinical trials, case series studies, letters to the editor, and studies that either failed to describe the prevalence of MAU or lacked pertinent data were also excluded.

Data Extraction

The information on the name of the primary author, publication year, country where the study was conducted, study design, sample size, types of diabetes, the overall prevalence of MAU, the prevalence of MAU between genders, and the prevalence of MAU between type 1 and type 2 were all summarised by Microsoft Excel as part of the data extraction format. Consensus-based talks and an independent evaluation by a third author were used to settle any differences or discrepancies. Consensus-based talks and an independent evaluation by a third researcher were used to settle any differences or disagreements. When more information was required, we tried to get in touch with the primary authors. For studies appearing in more than one publication, we considered the most recent and comprehensive studies and those with the largest sample size. In addition, studies conducted in both type 1 and type 2 DM subjects and reporting the prevalence of MAU independently were extracted as separate studies.

Quality Assessment

A Newcastle-Ottawa scale adapted for cross-sectional studies and a quality assessment tool were used to assess the quality of the included studies.³⁴ All eligible studies were evaluated, and studies of good quality or above were included in the final analysis. Two authors (OM and EA) evaluated each featured paper’s quality independently. Before determining the final evaluation score, the reviewers compared their quality appraisal scores and eliminated any discrepancies.

Operational Definitions

Patients with a significant rise in the excretion of urine albumin-creatinine ratio (ACR) within the particular range of 30–299 mg of albumin per g of creatinine or a urinary albumin excretion rate (UAE) of between 30–299 mg per 24 hours were regarded as having microalbuminuria. DM was defined as HbA1c (≥6.5%), a fasting blood sugar level (≥126 mg/dl), or a random blood sugar level (≥200 mg/dl).

Statistical Analysis

STATA^TM version 14 software was used for all data analysis. Using a random-effects model, a meta-analysis of the pooled prevalence of MAU was conducted, producing a pooled prevalence with 95% confidence intervals. Cochran’s Q and the I² statistics were used to assess the degree of heterogeneity between studies. With I² values of 25%, 50%, and 75%, respectively, the degree of heterogeneity is classified as low, moderate, or high.³⁵ Subgroup analyses were conducted in accordance with the predefined criteria, including sub-region, publication year, sex, and forms of DM, in order to investigate the origins of heterogeneity. The symmetry of the funnel plot was visually examined, and Egger’s test statistics were considered to evaluate the publication bias among the studies.^36,37 The random effect analysis was utilised to do a nonparametric trim and fill analysis because both techniques revealed publication bias. A leave-one-out sensitivity analysis was additionally conducted to assess the impact of single research on the combined estimate of the other studies.³⁸ For all computations, statistical significance was set at p <0.05 for all calculations.

Result

Flow Chart

Figure 1 shows the flow chart and selection process for determining the pooled prevalence of microalbuminuria (MAU) in diabetes patients. A total of 747 articles were discovered using electronic searches and other sources, and 521 non-duplicate articles were screened. About 483 of them were removed after looking over their titles and abstracts. The full texts of the remaining 38 articles were reviewed. Only 29 studies met our inclusion criteria, and the remaining 9 articles were rejected because the intended outcomes were not attained. Moreover, the four studies that reported the prevalence of MAU in type 1 DM and type 2 DM were extracted separately and treated as independent investigations. Subsequently, 29 articles were included in the final analysis (Figure 1).

Figure 1 Flow chart of studies’ search and retrieval process.

Overview of Included Studies

In this study, 29 original articles published from 1992 to 2022, consisting of 3885 study participants, were included.^39–67 Among the study participants, 824 (21.2%) had type 1 diabetes, 2343 (60.3%) had type 2 diabetes, and for the remaining 718 (18.5%), the specific types of DM were not reported. Eleven (33.3%) studies were conducted on type 1 DM patients, seventeen studies (51.5%) on type 2 DM patients, and five studies (15.2%) on type 1 and 2 DM patients. All the included studies were conducted at the health institution level, and no community-based study was found. Except for one case-control study,^44,61,66 and one cohort study,⁵⁸ all the studies used a cross-sectional study design with sample sizes ranging from 20⁶⁷ to 289 patients.⁵⁶ The mean age of the study participants varied from 8.4 to 61.3 years. Moreover, Omar et al⁶¹ reported the highest (77.5%) prevalence of MAU in type 1 DM patients, while Amolo⁶⁰ reported the lowest (6.2%) prevalence. For analysis purposes, studies done by Rahlenbeck et al, Martin et al, Lutale et al, and Fetni et al were extracted twice because they reported the prevalence of MAU separately for type 1 and type 2 DM. The studies were conducted in 14 different countries, including Ethiopia (2 studies),^40,41 Nigeria (6 studies),^{42–45,66,67} Cameroon (3 studies),^46,47,64 Sudan (2 studies),^48,65 Tanzania (3 studies),^49–51 Uganda (2 studies),^52–54 South Africa (1 study),⁵⁵ Botswana (2 studies),^56,57 Algeria (1 study),⁵⁸ Ghana (1 study),⁵⁹ Kenya (1 study),⁶⁰ Egypt (1 study),⁶¹ Senegal (1 study),⁶² and Zambia (2 studies)^39,63 (Table 1).

Table 1 Overview of Included Studies Conducted in Africa (N=3885)

Authors (Year) [Reference]	Country	Study Design	Setting of the Study	Mean Age (Year)	Sample Size	Types of DM	Comorbidity	Duration of DM in Years	Prevalence of MAU N (%)
Rasmussen et al (2013)^Citation39	Zambia	Cross-sectional	Hospital	NA	193	Type 1 and 2	Yes	NA	57 (29.4)
Rahlenbeck et al (1997)^Citation40	Ethiopia	Cross-sectional	Hospital	31.4±8.8	99	Type 1	Yes	6.0±4.9 years	32 (32)
Rahlenbeck et al (1997)^Citation40	Ethiopia	Cross-sectional	Hospital	56.7±11.3	71	Type 2	Yes	5.3±3.9 years	26 (37)
Muse et al (2020)^Citation41	Ethiopia	Cross-sectional	Hospital	55.9±13.3	204	Type 2	Yes	NA	98 (48)
Ogiator et al (2020)^Citation42	Nigeria	Cross-sectional	Hospital	53.12±11.66	93	Type 2	Yes	NA	33 (35.5)
Ufuoma et al (2016)^Citation43	Nigeria	Cross-sectional	Hospital	55.2 ± 8.5	200	Type 2	Yes	8.1±6.8 years	116 (58)
Halliru et al (2016)^Citation44	Nigeria	Case-control	Hospital	42± 1.8	100	Type 2	Yes	NA	34 (34)
Erasmus et al (1992)^Citation45	Nigeria	Cross-sectional	Hospital	51.1	113	Type 2	Yes	5.1 years	59 (52)
Bissong et al (2017)^Citation46	Cameroon	Cross-sectional	Hospital	NA	81	Type 2	Yes	NA	28 (34.6)
Efundem et al (2017)^Citation47	Cameroon	Cross-sectional	Hospital	55.3 ± 10.2	162	Type 2	Yes	6.3 ± 5.2	23 (14.2)
Rahamtalla et al (2012)^Citation48	Sudan	Cross-sectional	Hospital	NA	58	Type 2	Yes	NA	26 (44)
Lutale et al (2007)^Citation49	Tanzania	Cross-sectional	Hospital	NA	91	Type 1	Yes	3 years	11 (12)
Lutale et al (2007)^Citation49	Tanzania	Cross-sectional	Hospital	NA	153	Type 2	Yes	4 years	15 (9.8)
Ghosh et al (2012)^Citation50	Tanzania	Cross-sectional	Hospital	61.3±10.5	149	Type 2	Yes	8±7 years	43 (29)
Kantarama et al (2021)^Citation51	Tanzania	Cross-sectional	Hospital	NA	124	Type 2	Yes	NA	77 (62.1)
Martin et al (2018)^Citation52	Uganda	Cross-sectional	Hospital	46 ± 15	140	Type 2	Yes	2 months	64 (45.7)
Martin et al (2018)^Citation52	Uganda	Cross-sectional	Hospital	NA	35	Type 1	Yes	2 months	19 (54.3)
Kiconco et al (2019)^Citation53	Uganda	Cross-sectional	Hospital	NA	140	Type 1 and 2	Yes	6.8 years	32 (22.9)
Lubwama S (2022)^Citation54	Uganda	Cross-sectional	Hospital	8.4	153	Type 1	Yes	4.35 years	21 (13.7)
Kalk et al (2010)^Citation55	South Africa	Cross-sectional	Hospital	34.9±8.6	68	Type 1	Yes	8 years	27 (39.7)
Molefe Baikai et al (2018)^Citation56	Botswana	Cross-sectional	Hospital	52	289	Type 2	Yes	NA	129 (44.6)
Ramaphane et al (2021)^Citation57	Botswana	Cross-sectional	Hospital	18.7 ±5	127	Type 1	Yes	6.6 ±4.6 years	36 (28.3)
Fetni et al (2021)^Citation58	Algeria	Cohort	Hospital	55±15.9	252	Type 2	Yes	NA	87 (34.5)
Fetni et al (2021)^Citation58	Algeria	Cohort	Hospital	20	40	Type 1	Yes	NA	16 (40)
Eghan et al (2007)^Citation59	Ghana	Cross-sectional	Hospital	54.2±10.9	109	Type 2	Yes	11.96 ±7.4 years	47 (43.1)
Amolo P (2010)^Citation60	Kenya	Cross-sectional	Hospital	10.9±4.5	65	Type 1	Yes	NA	4 (6.2)
Omar et al (2010)^Citation61	Egypt	Case-control	Hospital	11.78±3.83	40	Type 1	Yes	6.77±3.1 years	31 (77.5)
Djiby et al (2018)^Citation62	Senegal	Cross-sectional	Hospital	56.6±8.0	221	Type 1 and 2	Yes	NA	63 (28.5)
Chaamba S (2019)^Citation63	Zambia	Case-control	Hospital	52±13.6	45	Type 2	Yes	NA	28 (62.2)
Sobngwi et al (1999)^Citation64	Cameroon	Cross-sectional	Hospital	52.491.8	64	Type 1 and 2	Yes	5.8±90.8 years	34 (53.1)
Mohamed A (2005)^Citation65	Sudan	Cross-sectional	Hospital	13.6±2.6	86	Type 1	Yes	4.4±3.4 years	33 (38.4)
Bunza et al (2014)^Citation66	Nigeria	Case-control	Hospital	NA	100	Type 1 and 2	Yes	5.33 years	22 (22)
Yarhere et al (2020)^Citation67	Nigeria	Cross-sectional	Hospital	13.3± 3.7	20	Type 1	Yes	2.9 ± 0.44 years	12 (60)

Abbreviation: NA, not available.

The Pooled Prevalence of Microalbuminuria Among Diabetes Patients in Africa

The pooled prevalence of microalbuminuria in the studies conducted in Africa between 1992 and July 4, 2022, was found to be 37.11% (95% CI 31.27–42.95; p < 0.001). The Cochran’s Q values for the heterogeneity test are 569.22 (degree of freedom, d.f = 32), and I² with 94.4% suggests that there is a significant amount of heterogeneity between studies (Figure 2).

Figure 2 Forest plot showing the pooled prevalence of microalbuminuria among diabetes patients in Africa.

Subgroup Analysis

Subgroup analysis was done based on sub-region, types of DM, mean duration of DM, study design, sex, and publication year in order to investigate the cause of heterogeneity among the included studies in this meta-analysis. Eastern Africa accounted for more than one-third (12 (36.4%) of the studies in our review, followed by Western Africa (8 (24.2%), Central Africa (3 (9.1%), Northern Africa (5 (15.2%), and Southern Africa (5 (15.2%). In five African sub-regions, the combined prevalence of MAU among DM patients ranged from 30.57% (95% CI: 20.3–40.83) in Eastern Africa to 46.5% (95% CI: 32.29–60.71) in Northern Africa. The prevalence estimates between studies per sub-region showed significant heterogeneity (total heterogeneity, p<0.0001) (Table 2). 33.95 (26.83, 41.08).

Table 2 The results of Subgroup Analysis by Different Categories of the Studies in Africa

Subgroup	Category	No. of Studies	Prevalence of MAU (95% CI)	P-value	I^2 (%)
Sub-region	Eastern	12	30.57 (20.3, 40.83)	<0.001	96.0
	Western	8	40.81 (30.85, 50.76)	<0.001	90.3
	Central	3	33.47 (10.46, 56.48)	<0.001	94.8
	Northern	5	46.50 (32.29, 60.71)	<0.001	88.8
	Southern	5	39.79 (29.81, 49.77)	<0.001	86.3
Types of DM	Type 1	11	35.34 (23.89, 46.80)	<0.001	94.2
	Type 2	17	40.24 (32.00, 48.47)	<0.001	94.9
	Type 1 and 2	5	30.02 (22.41, 37.63)	<0.001	80.5
Sex	Male	23	36.6 (29.50, 43.69)	<0.001	87.4
	Female	23	33.95 (26.83, 41.08)	<0.001	91.4
Year of publication	1992–2002	4	43.34 (32.61, 54.08)	0.005	76.7
	2003–2012	8	27.11 (16.37, 37.85)	<0.001	93.7
	2013–2022	21	39.68 (32.71, 46.65)	<0.001	93.9
Duration of diabetes	< 5 years	7	31.48 (18.73, 44.23)	<0.001	94.4
	> 5 years	13	38.73 (29.34, 48.13)	<0.001	93.8
Study design	Cross-sectional	27	35.56 (29.12, 42.01)	<0.001	94.7
	Case-control	4	48.46 (24.14, 72.78)	<0.001	95.2
	Cohort	2	35.22 (29.74, 40.69)	0.51	0.0

Furthermore, this meta-analysis revealed that the overall pooled estimates of MAU among type 1 and type 2 DM patients in Africa were 35.34% (95% CI: 23.89–46.80, I²=94.2), and 40.24% (95% CI: 32.0–48.47, I²=94.9), respectively. Based on the subgroup analysis of MAU, the pooled point estimates for diabetic male patients were 36.6% (95% CI: 29.50–43.69, I²=87.4), and diabetic female patients were 33.95% (95% CI: 26.83–41.08, I²=91.4). High levels of heterogeneity were seen in both sexes (p<0.0001) (Table 2). In order to see the prevalence of MAU over time, we divided the included studies into three groups (1992–2002, 2003–2012, and 2013–2022), or every ten years intervals, based on the publication year. Among studies conducted between 1992 and 2002, a pooled prevalence of 43.3% (95% CI: 32.6–54.1, I²=76.7) of MAU was found. The lowest prevalence of MAU, however, was observed in studies conducted from 2003 to 2012, when it was 27.11% (95% CI: 16.37–37.85, I²=93.7). MAU, on the other hand, was more common in people with diabetes for more than 5 years 38.73% (95% CI: 29.34–48.13) than in people with diabetes for less than 5 years 31.48% (95% CI: 18.73–44.23). Moreover, the subgroup analysis revealed that among case-control studies, the highest pooled prevalence of MAU was found to be 48.46% (95% CI: 24.14–72.78) (Table 2).

Publication Bias

A funnel plot test was used to assess the presence of publication bias. And this showed the presence of publication bias, which is supported by the asymmetry displayed in the funnel plot. Moreover, Egger’s test statistics also indicated the presence of publication bias with a p-value of <0.001. Therefore, this is an indication of the presence of unpublished data that can modify the prevalence of MAU among DM patients in Africa (Figure 3).

Figure 3 Bias assessment plot of reported prevalence of microalbuminuria among diabetes patients across studies published in Africa.

Trim and Fill Analysis of Pooled Prevalence of Microalbuminuria Among Diabetes Patients

The pooled prevalence of microalbuminuria among DM patients in Africa was 23.94% (95% CI: 17.56–30.33) based on trim and fill analysis following the addition of fourteen studies, with a p-value <0.001 (Table 3).

Table 3 Trim and Fill Analysis of Overall Pooled Prevalence of Microalbuminuria Among Diabetes Patients in Africa

Meta-Analysis
Method	Pooled Est.	95% CI	Z-value	p-value		No. of Studies
Fixed	30.78	29.42–32.14	44.50	<0.001		33
Random	37.11	31.27–42.95	12.45	<0.001
Iteration	Estimate	Tn	# to Trim		Diff
1	30.78	406	8		561
2	26.57	459	11		106
3	23.41	492	13		66
4	22.50	502	14		20
5	22.31	503	14		2
6	21.31	503	14		0
Filled meta-analysis
Method	Pooled Est.	95% CI	Z-value	p-value		No. of Studies
Fixed	22.31	21.11–23.50	36.66	<0.001		47
Random	23.94	17.56–30.33	7.35	<0.001

Notes: Test for heterogeneity: Q= 569.22 on 32 degrees of freedom (p= 0.000). Moment-based estimate of between studies variance = 268.38. Trimming estimator: Linear. Meta-analysis type: Fixed-effects model. Test for heterogeneity: Q= 1276.74 on 46 degrees of freedom (p= 0.000). Moment-based estimate of between studies variance = 470.26.

Sensitivity Analysis

Sensitivity analysis was employed to determine the impact of a single study on the combined effect size. A single study’s exclusion had no discernible impact on the pooled burden estimates when studies were eliminated one at a time since the resulting pooled effect size was within the 95% confidence interval of the combined pooled effect size. Because of this, random-effects sensitivity analysis revealed that no particular study had an effect on the overall prevalence of MAU among diabetic patients (Table 4).

Table 4 Sensitivity Analysis for Single Study Influence of Pooled Estimate

S.No	Study Omitted	Estimate	95% CI
1	Rahlenbeck et al (1997)^Citation40	37.28	31.28–43.29
2	Rahlenbeck et al (1997)^Citation40	37.12	31.15–43.09
3	Muse et al (2020)^Citation41	36.75	30.83–42.67
4	Ogiator et al (2020)^Citation42	37.17	31.18–43.16
5	Ufuoma et al (2016)^Citation43	36.37	30.66–42.09
6	Halliru et al (2016)^Citation44	37.22	31.22–43.22
7	Erasmus et al (1992)^Citation45	36.63	30.74–42.52
8	Bissong et al (2017)^Citation46	37.20	31.21–43.18
9	Efundem et al (2017)^Citation47	37.87	31.95–43.79
10	Rahamtalla et al (2012)^Citation48	36.91	30.97–42.85
11	Lutale et al (2007)^Citation49	37.92	32.03–43.82
12	Lutale et al (2007)^Citation49	37.98	32.25–43.71
13	Ghosh et al (2012)^Citation50	37.39	31.35–43.43
14	Kantarama et al (2021)^Citation51	36.27	30.53–42.0
15	Martin et al (2018)^Citation52	36.84	30.89–42.79
16	Martin et al (2018)^Citation52	36.64	30.74–42.55
17	Kiconco et al (2019)^Citation53	37.59	31.56–43.62
18	Lubwama et al (2022)^Citation54	37.88	31.97–43.79
19	Kalk et al (2010)^Citation55	37.04	31.08–43.0
20	Molefe et al (2018)^Citation56	36.87	30.89–42.85
21	Ramaphane et al (2021)^Citation57	37.41	31.39–43.44
22	Fetni et al (2021)^Citation58	37.23	31.14–43.32
23	Fetni et al (2021)^Citation58	37.03	31.09–42.98
24	Eghan et al (2007)^Citation59	36.93	30.96–42.90
25	Amolo P (2010)^Citation60	38.09	32.37–43.80
26	Omar et al (2014)^Citation61	35.88	30.18–41.58
27	Djiby et al (2018)^Citation62	37.42	31.33–43.52
28	Chaamba S (2019)^Citation63	36.38	30.52–42.24
29	Sobngwi et al (1999)^Citation64	36.63	30.73–42.53
30	Mohamed A (2005)^Citation65	37.08	31.10–43.05
31	Bunza et al (2014)^Citation66	37.61	31.61–43.60
32	Yarhere et al (2020)^Citation67	36.57	30.68–42.46
33	Rasmussen et al (2013)^Citation39	37.39	31.32–43.46
	Combined	37.11	31.27–42.95

Discussion

End-stage renal disease has become more common as the prevalence of diabetes mellitus has increased.⁶⁸ Microalbuminuria is a risk factor for cardiovascular disease complications and an early indicator of diabetic nephropathy.²⁰ The presence of microalbumin in diabetic patients’ urine is an early warning sign of systemic vasculopathy and other microvascular complications.⁶⁹ According to the current study, the overall pooled prevalence of MAU among diabetes patients in Africa was 37.11% (95% CI: 31.27–42.95). This finding indicated that a significant proportion of diabetes patients had microalbuminuria. Furthermore, the current study discovered that 35.34% (95% CI: 23.89–46.80) of type 1 diabetes patients and 40.24% (95% CI: 32.0–48.47) of type 2 diabetes patients have microalbuminuria. This high prevalence of MAU in African diabetics could be attributed to the presence of comorbidities such as hypertension, which exacerbates systemic vasculopathy and other microvascular complications.

This pooled prevalence of MAU among type 2 diabetes patients is comparable to the findings (39%) of a global cross-sectional study on type 2 diabetes patients conducted in 33 countries.⁷⁰ A similar result of 35.1% (12.3–74.5%) was reported in large, multiple international cohort studies including >3 million participants.⁷¹ Likewise, the current pooled prevalence of microalbuminuria among diabetic subjects was in agreement with the Pakistan multi-centre study (34%),⁷² Albania (40.8%),⁷³ Asians (39.8%),⁷⁰ and Saudi Arabians (41.3%).⁷⁴ Furthermore, Dinneen et al discovered that the prevalence of microalbuminuria in type 2 diabetes mellitus patients ranged from 20% to 36% across 8 cohorts.⁷⁵ However, our finding was higher than the European prevalence of microalbuminuria (26% to 29%),⁷⁶ Australia (26.1%),⁷⁷ North India (25.5%),⁷⁸ and Iran (14.2%).⁷⁹ Moreover, Klein et al found a prevalence of microalbuminuria of 22.0% in those type 2 DM subjects.⁸⁰ On the other hand, the MAU in our study was lower than a study conducted by Parving et al (58%)⁸¹ and the United Arab Emirates (61%).⁸²

The current overall estimated prevalence of MAU in type 1 diabetes was 35.34%, which was comparable to a study done by Klein et al (29.2%).⁸⁰ In contrast, the current result was lower than that reported by Warram et al (60%).⁸³ Changes in sample size, study design, data processing method, assay method, ethnicity, cut-off levels for albumin excretion, duration of DM, renal status, and other clinical features could explain this disparity. Furthermore, the prevalence of microalbuminuria was heavily influenced by the variability and concentration of urine on a daily basis. According to this review’s evidence, more than one-third of diabetes patients have microalbuminuria, which is the first step in the development of microvascular complications such as renal nephropathy.²⁰ Renal function deteriorates as a result of diabetic nephropathy, resulting in renal insufficiency. Treatment is required at this stage to slow the rate of progression. If left untreated, the kidney can malfunction, leading to kidney failure and the need for dialysis or kidney transplants,^84,85 but such kidney disease treatments and management are difficult to come by in resource-limited countries.^32,86

The present review also attempted to conduct a subgroup analysis based on sub-region, gender, publication year, and diabetes type. MAU had the highest pooled prevalence of 46.50% (95% CI: 32.29–60.71) in studies conducted in Northern Africa,^48,58,61,65 followed by 40.81 (95 CI: 30.85–50.76) in studies conducted in Western Africa.^{42–45,59,62,66,67} This disparity could be attributed to differences in study subjects, urine collection methods, the presence of comorbidity, and assay methods used in those African countries. In terms of gender, the subgroup analysis revealed that 36.60% (95% CI: 29.50–43.69) of male DM patients had MAU compared to 33.95% (95% CI: 26.83–41.08) of female DM patients. In previous studies,^87,88 male participants had a higher proportion of MAU than their female counterparts. This slightly higher prevalence of MAU in male subjects could be explained by males being more vulnerable to other risk factors that could aggravate diabetes and affect renal glomerulus function.

In the current review, 40.24% of type 2 diabetic patients had MAU, which was higher than that of type 1 diabetes patients by 5%. In addition, this review also indicated that the prevalence of MAU among studies conducted on type 1 and type 2 DM was even lower (30.02%). Albuminuria may increase in type 2 diabetes patients due to other comorbidities, and renal function, particularly glomerular filtration rate, may decrease. Studies published twenty years ago^40,45,64 revealed a higher prevalence (43.3%) of MAU than recent ones. Even after a subgroup analysis, this review found significant heterogeneity in studies on the prevalence of microalbuminuria in DM patients. There are several potential causes of heterogeneity in this study. One possible explanation is sample size differences; for example, Yarhere et al⁶⁷ included only 20 participants in their study, whereas Molefe Baikai et al⁵⁶ included 289 participants. Another possible explanation for this heterogeneity is the variety of urine sample collection methods, assay methods, and differences in diabetes duration among participants. This review also revealed a publication bias, which is supported by the symmetry of the funnel plot and the results of Egger’s test. After adding 14 studies, we performed trim and fill analysis, and the pooled prevalence of microalbuminuria among DM patients was found to be 23.94%. The sensitivity analysis, however, revealed that no single study had an effect on the total pooled effect size.

One of the current review’s limitations is that it was unable to provide information on the risk factors associated with MAU. However, this is the first review to combine the findings of several African studies, providing stronger evidence on the prevalence of microalbuminuria among diabetes patients. Furthermore, the review attempted to conduct a subgroup analysis based on diabetes type and demonstrated the pooled prevalence of MAU among type 1 and 2 DM patients.

Conclusion

This review discovered a high prevalence of microalbuminuria among diabetes patients in Africa. Microalbuminuria was more common in Northern Africans, people with type 2 diabetes, and people who had diabetes for more than five years. As a result, early detection of microalbuminuria is critical for preventing and treating microvascular complications such as diabetic nephropathy and the onset of end-stage renal disease. This is especially important in developing countries, where recommended renal therapies such as dialysis and transplantation are not easily accessible due to financial constraints.

Data Sharing Statement

All the datasets used and/or analyzed during the current study are available in the manuscript.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors declare that they have no potential competing interests in this work.

Additional information

Funding

There is no funding to report.