Chronic kidney disease hypertension diabetes and obesity in the adult population of Morocco: how to avoid “over”- and “under”-diagnosis of CKD

Chronic kidney disease, hypertension, diabetes, and obesity in the adult population of Morocco:

how to avoid “ over ” - and “ under ” -diagnosis of CKD

Mohammed Benghanem Gharbi ^1,6 , Monique Elseviers ^2,6 , Mohamed Zamd ¹ , Abdelali Belghiti Alaoui ³ , Naı¨ma Benahadi ³ , El Hassane Trabelssi ³ , Rabia Bayahia ⁴ , Benyoune`s Ramdani ¹ and Marc E. De Broe ^5,6

1

Faculty of Medicine and Pharmacy, University Hassan II, Casablanca, Morocco;

Department of Biostatistics, Center for Research and Innovation in Care, University of Antwerp, Antwerp, Belgium;

Ministry of Health, Rabat, Morocco;

Faculty of Medicine and Pharmacy, University Mohammed V, Rabat, Morocco; and

University of Antwerp, Antwerp, Belgium

The prevalence of hypertension, diabetes, obesity, and chronic kidney disease (CKD) in an adult Arabic-Berber population was investigated according to 2012 KDIGO guidelines. A stratified, randomized, representative

sample of 10,524 participants was obtained. Weight, height, blood pressure, proteinuria (dipstick), plasma creatinine, estimated glomerular fi ltration rate, and fasting glycemia were measured. Abnormal results were controlled within 2 weeks; eGFR was retested at 3, 6, and 12 months. The population adjusted prevalences were 16.7% hypertension, 23.2% obesity, 13.8% glycemia, 1.6% for eGFR under 60 ml/min/1.73 m

and con fi rmed proteinuria 1.9% and hematuria 3.4%. Adjusted prevalence of CKD was 5.1%;

distribution over KDIGO stages: CKD1: 17.8%; CKD2: 17.2%;

CKD3: 52.5% (3A: 40.2%; 3B: 12.3%); CKD4: 4.4%; CKD5:

7.2%. An eGFR distribution within the sex and age categories was constructed using the third percentile as threshold for decreased eGFR. A single threshold (under 60 ml/min/1.73 m

) eGFR classifying CKD3 – 5 leads to

“ overdiagnosis ” of CKD3A in the elderly, overt

“ underdiagnosis ” in younger individuals with eGFR over 60 ml/min/1.73 m

, below the third percentile, and no proteinuria. By using the KDIGO guidelines in a correct way, “ kidney damage ” (con fi rmed proteinuria, hematuria) and the demonstration of chronicity of decreased eGFR

< 60 ml/min/1.73 m

, combined with the third percentile as a cutoff for the normality of eGFR for age and sex, overcome false positives and negatives, substantially decrease CKD3A prevalence, and greatly increase the accuracy of identifying CKD.

Kidney International (2016)

,

–

; http://dx.doi.org/10.1016/

j.kint.2016.02.019

KEYWORDS: chronic kidney disease; diabetes; hypertension; Morocco;

obesity; overdiagnosis of CKD; underdiagnosis of CKD

Copyright ª 2016, International Society of Nephrology. Published by Elsevier Inc. All rights reserved.

O ver the past 15 years, chronic kidney disease (CKD) received scienti fi c attention as an important public health problem.

Although the incidence of end- stage renal disease remains low, the cost of renal replacement therapy is among the highest per patient of any form of chronic disease.

In many countries with low- to medium- sized economies, transplantation/dialysis facilities are unaf- fordable.

In a recent survey in Africa, it was found that the accessibility to renal replacement therapy varied between 100% mainly in North Africa decreasing to <10% in several sub-Saharan countries (M. Benghanem Gharbi, personal communication, 2012).

Emphasis on early detection of CKD and secondary prevention, by controlling several well-known risk factors and the use of low-cost medication are potential cost-effective solutions for developing countries.

In emerging countries, reliable information on the preva- lence of CKD is unavailable. In addition, studies from the developed world suffer from major flaws,

such as lack of randomized sample, no confirmation of proteinuria, no demonstration of chronicity of estimated glomerular filtra- tion rate (eGFR) <60 ml/min/1.73 m

,

and assay problems for serum creatinine. These factors and others lead to the overreporting of CKD prevalence.

In Morocco, information on the prevalence of the different stages of CKD was unknown. The MAREMAR (Maladie Rénale Chronique au Maroc) study, a screening and preven- tion program of CKD, is a joint venture including the Moroccan health authorities, International Society of Nephrology, Moroccan Society of Nephrology and World Health Organization. This paper reports the cross-sectional baseline observations on CKD, hypertension, diabetes, and obesity in this country.

RESULTS Recruitment

The recruitment lasted 6 months in Khemisset (December 2009–June 2010), 10 months in El Jadida (May 2010–March 2011) including a 3-month break due to the summer holiday period and Ramadan. Only 50.5% of subjects selected from the initial list were found. They were replaced by participants of the reserve list. The response rates showed that 85.02% of

Correspondence: M.E. De Broe, University of Antwerp, Laboratory of Pathophysiology, Universiteitsplein 1, B-2610 Wilrijk, Belgium. E-mail:

marc.debroe@uantwerpen.be

6

These authors contributed equally to this work.

Received 14 December 2015; revised 28 January 2016; accepted 11

February 2016

the found/visited participants agreed to join the study. Once agreed, 82.56% of the latter received the baseline screening.

Almost all (97.7%) participated in the second con fi rmation visit. Baseline screening was performed in 10,524 participants, equally spread over the predefined sex and age strata (Supplementary Figure S1). A description of the study pop- ulation is shown in Table 1.

Renal function

Mean serum creatinine was 8.1 4.3 mg/l. Mean eGFR was 103.0 26.8 ml/min/1.73 m

. The adjusted prevalence of

eGFR <60 ml/min/1.73 m

was 1.6% (95% confidence

interval [CI]: 1.4–1.8). Figure 1a shows the distribution of eGFR within the 3 age categories. Figure 1b shows the percentiles of eGFR distribution within the sex and age categories.

Chronicity of decreased eGFR was investigated in 78.9% of the subjects (n ¼ 285) with CKD3A, 3B, 4, and 5. The remaining were deceased or lost to follow-up. The majority (75%) of false positives were found in the subjects with CKD3A. Thirty-two percent of the CKD3A subjects and 7.4% of the CKD3B subjects had an eGFR > 60 ml/min/

1.73 m

when reinvestigated after 3 months or longer.

Subjects with CKD4 and 5 ( n ¼ 51) remained in these low eGFR categories, and 11 were on dialysis, died, or lost to follow-up after 3 months or longer.

Proteinuria

The dipstick analysis at the fi rst visit resulted in proteinuria in 513 subjects (4.9%): mild ( þ ) in 408 (79.5%), and overt

( >þ ) in 105 (20.5%). Of these 513, a second dipstick

investigation was performed in 384 subjects including 69.3% of subjects with mild proteinuria and all subjects with overt proteinuria. The prevalence of confirmed proteinuria (extrapolating the results obtained in 69.3% of subjects with mild proteinuria) was 206 (1.9%), mild in 139 cases and overt in 67 (36þþ, 25þþþ, 6þþþþ). A false-positive result was found in 67.5% of subjects with mild proteinuria and decreased substantially to 28.7% in overt proteinuria. The distribution of the mild and overt proteinuria among the different KDIGO (Kidney Disease: Improving Global Outcomes)-CKD stages is shown in Figure 2a.

Hematuria

The dipstick analysis at the fi rst visit resulted in hematuria ($þ) in 921 subjects (8.8%). Of these 921, a second dipstick investigation was performed in 795 subjects showing hema- turia in 359 (45.2%) indicating a false-positive result in 54.8%. The prevalence of confirmed hematuria (extrapolating the results obtained from 86.3%) was 3.4%. The prevalence of hematuria was 2.4% (n ¼ 130) in men and 4.5% (229) in women. The distribution of hematuria among the different KDIGO-CKD stages is shown in Figure 2b.

Chronic kidney disease

When eGFR and confirmed proteinuria were taken into account, the CKD1 and 2 KDIGO classes represented, respectively, 17.8% and 17.2% of the total CKD population (n ¼ 442). CKD3 represented the most prevalent stage:

52.5% (3A: 40.2%; 3B: 12.3%), whereas CKD4 and 5 repre- sented, respectively, 4.4% and 7.2% of the total CKD popu- lation. The majority of the subjects with CKD3A (79.2%) and 3B (58.1%) had no proteinuria nor hematuria, whereas in subjects with CKD5, 45% still had no proteinuria nor hematuria (Figure 2b).

The majority of male (78.2%) and female (65.2%) subjects with CKD class 3A (eGFR 60 – 45 ml/min/1.73 m

) fall within the 3rd to 10th percentiles of eGFR of their age and sex (Figure 3a and b). Six men (15.6%) and no women in class CKD3B, and not a single subject in classes CKD4 and 5 were observed with eGFR above the 3 percentile values of eGFR of their age/sex (Figure 3a and b). There were twice as many men (n ¼ 21) as women (n ¼ 11) with CKD5.

The dots in the green-backdrop zone of Figure 3a and b represent 130 young individuals (1.2% of study population).

The majority were younger than 45 years of age, had an eGFR

> 60 ml/min/1.73 m

, and were situated clearly below the third percentile of the distribution of the normal values of eGFR of their age/sex categories. No proteinuria and no hematuria was found in 96% of the cases. These subjects, however, had a decreased eGFR for their age/sex and, nevertheless, could not be classi fi ed to any of the KDIGO-CKD stages.

In view of the important variety in reported CKD preva- lences, mostly based on incomplete application of KDIGO guidelines, the data of the current study were analyzed using several combinations. A decrease of almost 50% in the preva- lence of CKD was noted when confirmation of “proteinuria/

hematuria and chronicity” of eGFR were included (Table 2).

Hypertension

The prevalence of confirmed hypertension was 21.9% with 7.2%, 20.3% and 39.2% for the 3 age categories of 26 to 40, 41 to 55, and 56 to 70 years, respectively. The adjusted prevalence of hypertension was 16.7% (Table 3). In subjects without hypertension, 0.8% had confirmed mild and 0.3% confirmed overt proteinuria. In contrast in those with hypertension, 1.6% had confirmed mild and 1.7% had confirmed overt proteinuria.

Table 1 | Description of the study population

n (%)

Number of participants 10,524 (100)

Town of recruitment

Khemisset 5467 (51.9)

El Jadida 5057 (48.1)

Age categories, yr

26 – 40 3506 (33.3)

41 – 55 3689 (35.1)

56 – 70 3329 (31.6)

Sex

Male 5122 (48.7)

Female 5402 (51.3)

c l i n i c a l i n v e s t i g a t i o n M Benghanem Gharbi et al.: MAREMAR study

Figure 1 | Distribution of eGFR by age and age/sex. (a) Distribution of estimated glomerular fi ltration rate (eGFR) within the 3 age categories.

The dark vertical lines represent from left to right 15, 30, 60, and 90 ml/min/1.73 m

of eGFR. The changes in renal function (eGFR, MDRD

[Modi fi cation of Diet in Renal Disease] formula) in the 3 age classes are distributed in a Gaussian way with a shift of the curve to the left in the

aging population. (b) The eGFR distribution showing the 3rd, 10th, 25th, 50th, 75th, 90th, and 97th percentiles (P) within the sex and age

categories (n ¼ 10,524). The “ normal ” decline in eGFR of the study population is 0.75 ml/min/1.73 m

per year.

Figure 2 | KDIGO classi fi cation based on eGFR, proteinuria, and hematuria. (a) KDIGO (Kidney Disease: Improving Global Outcomes) classi fi cation based on estimated glomerular fi ltration rate (eGFR) and con fi rmed proteinuria (pr) (mild, overt, dipstick investigation). (b) KDIGO classi fi cation based on eGFR, proteinuria, and hematuria (hemat) (con fi rmed dipstick investigation).

c l i n i c a l i n v e s t i g a t i o n M Benghanem Gharbi et al.: MAREMAR study

Obesity

Mean body weight was 74.6 kg ( 13.1) in men and 73.8 kg ( 13.9) in women. Mean body mass index was 24.9 ( 4.1) in men and 28.6 ( 5.2) in women. A body mass index >30 was observed in 24.2% of the subjects, in 11.3% of men and in 36.4% of women, and was already present in 27.4% of women of 26 to 40 years of age, increasing to >40% in older age groups of women. The adjusted prevalence of obesity (body mass index > 30) was 23.2% (Table 3). No statistical signi fi cant difference was found ( P ¼ 0.482) in the prevalence of CKD in the 5 subclasses of body mass index: < 20; 20 to 24;

25 to 29; 30 to 34; and > 30. There was no statistical difference between the prevalence of obesity in CKD3 subjects with or without confirmed proteinuria.

Glycemia

The mean glycemia was 1.13 0.49 g/l. The prevalence of glycemia $1.26 g/l was 6.3%, 18.3%, and 27.4% for the categories 26 to 40, 41 to 55, and 56 to 70 years of age, respectively. The adjusted prevalence of glycemia $1.26 g/l was 13.4% (Table 3).

DISCUSSION

MAREMAR is one of only a few epidemiological studies applying all criteria,

necessary to obtain precise prevalence numbers and classify correctly a subject to 1 of the 6 KDIGO- CKD classes. The current study used a randomized sample, had an 85% response rate, and had a relative short screening period, minimizing potential serum creatinine assay prob- lems. Also confirmation of proteinuria, hematuria, and chronicity of decreased eGFR found at the first screening were obtained. This is in contrast to most published studies.

The visit of a well-trained nurse explaining in detail the aims of the study, the personal bene fi t the participants could expect of their participation (follow-up, free medication for 5 years), and the visual simple representation of the different steps of this study in a booklet (Supplementary Figure S2) may have contributed to the high participation. The latter was decided on because the illiteracy rate of the Moroccan population is up to 30%.

Prevalence of CKD in Morocco is among the lowest in the world (5.1%), compared with other studies reporting preva- lences between 8.8% and 11.9%.

In our study, the adjusted prevalence of subjects with eGFR <60 ml/min/

1.73 m

was 1.6%; again a low percentage compared with the majority of the reported studies (prevalences between 1.7%

and 8.1%).

Figure 3 | Subjects with KDIGO CKD3 to 5 without/with proteinuria and or hematuria and those subjects who cannot be allocated to any of the KDIGO-CKD stages. (a) Male and (b) female subjects with KDIGO (Kidney Disease: Improving Global Outcomes) chronic kidney disease (CKD) stages 3 to 5 without and with con fi rmed mild, overt proteinuria and/or hematuria and the subjects (green backdrop) who cannot be allocated to any of the KDIGO-CKD stages (estimated glomerular fi ltration rate [eGFR] > 60 ml/min/1.73 m

, below the 3rd percentile line of normal age-sex eGFR distribution and without proteinuria/hematuria).

Table 2 | Assessment of CKD based on eGFR, proteinuria, hematuria, using MAREMAR data

Proteinuria (%)

Proteinuria/

hematuria (%) Single eGFR þ uncon fi rmed

proteinuria/hematuria (as in most studies)

7.0 14.0

Single eGFR þ con fi rmed proteinuria/

hematuria (as in some studies)

4.2 7.4

Chronicity eGFR þ con fi rmed proteinuria/hematuria (according to KDIGO 2012) (as in this study)

3.9 6.7

3 þ correction for total population 2.9 5.1

Mild to overt proteinuria at

rst visit (n

513): 4.9% of study population. Con

rmed mild to overt proteinuria (n

206): 1.9% of study population. False-positive result:

67.2% mild proteinuria/28.7% overt.

CKD, chronic kidney disease; eGFR, estimated glomerular

ltration rate; KDIGO,

Kidney Disease: Improving Global Outcomes; MAREMAR, Maladie Rénale Chronique

au Maroc.

The population pyramid of Morocco is tagine-like (broad base, small top), in contrast to most of the constrictive population pyramids used in CKD prevalence studies, having a broad dome aspect. This results in a low percentage of individuals aged more than 60 years, hence a low percentage of CKD3, the most prevalent CKD stage.

The second and one of most important reasons of “over- diagnosing” in most studies is due to limiting investigation of proteinuria to a 1-time event.

We and others found a high percentage (Table 2) of false-positive dipstick (þ) urine even with dipstick ( þþ ) to ( þþþ ).

The methodology used in the present study was the validated dipstick analysis of fresh morning mid-stream urine samples.

Many factors, however, can in fl uence protein excretion such as obesity, age, sex, distant in fl ammation, high blood pressure, infection, and drug use,

resulting in wide fl uctuations, hence false posi- tivity of proteinuria.

A third reason could be the MDRD (Modification of Diet in Renal Disease) study equation systematically under- estimating true renal function in healthy individuals and imprecision at higher levels of GFR.

A fourth and also important reason for the overdiagnosis is the absence in many, if not all, studies of the demonstration of chronicity of the decreased eGFR.

Twenty-five to 30% of the subjects initially categorized as stage 3 CKD will subsequently no longer fall into this category when repeated measurements are obtained over a subsequent time period (minimum 3 months),

(32% in our CKD3A cases). Our study, the study of Eriksen and Ingebretsen,

and a very recent one by Inker et al. ,

all indicate the utmost importance of proving chro- nicity of the initial decreased eGFR.

The prevalence rates of CKD4 and 5 are many times lower than that of CKD3, the most prevalent stage, consisting mainly of aged subjects.

These striking differences are compatible with the observation that the majority of the stage CKD3A subjects seldom progress to end-stage renal disease, putting the designation of CKD at that stage into question.

Such inaccurate CKD3A labeling in older individuals without proteinuria/hematuria (85.9% this study) and/or

hypertension (61.3% this study) has many undesirable effects such as unnecessary anxiety, unneeded additional in- vestigations, and even loss of insurability.

If a single arbi- trary threshold of eGFR (<60 ml/ml/1.73 m

) uncalibrated for age and sex, and not considering the presence or absence of proteinuria, is used in the definition of CKD3, huge vari- ations in cardiovascular risk and risk to develop progressive renal damage in CKD3 will occur particularly in the older population. There is no evidence that detection of CKD3A as such without investigating proteinuria/hematuria, partic- ularly in the elderly, will be of any diagnostic/therapeutic value.

At least 33% up to 50% of the people who meet the recent de fi nition of CKD are classi fi ed as stage 3A.

The number that will progress to end-stage renal disease was estimated (measured) at only 0.15% to 0.20% per year over 10 to 25 years.

The vast majority is older than 60 years of age, and many (72.2% current study) have an eGFR that falls within the normal range (3rd to 97th percentile) for their age and sex. Most of them have no confirmed proteinuria/

hematuria (Figure 2a and b). In an excellent study

in CKD3 patients, the 10-year cumulative incidence of renal failure (dialysis, transplantation) was 0.04 (95% CI: 0.03–0.06) in contrast to mortality of 0.51 (95 CI: 0.48–0.55) mainly due to cardiovascular diseases. Hence high mortality preempted apparently the development of renal failure in CKD3, particularly 3A patients. It has been shown

that CKD3 class, particularly CKD3A, patients have no additional risk of mortality when compared with similar aged individuals with an eGFR > 60 ml/min/1.73 m

after adjustment for age, race, sex, and comorbidities.

The PREVEND (Prevention of Renal and Vascular Endstage Disease) and other studies

have demonstrated that there is little increased risk of car- diovascular disease, either deaths or events, in stage CKD3A patients in the absence of proteinuria when compared with those with an eGFR of >60 ml/min/1.73 m

without pro- teinuria. Recently, de Jong and Gansevoort,

proposed that there is a need to improve the definitions of stage CKD3.

Confirmed proteinuria should be included in stage 3 before labeling a subject as CKD3.

Table 3 | Prevalences according to age categories and adult population of Morocco Observed prevalence (%) Observed prevalence according to age (%)

Adjusted prevalence according to the Moroccan population

N [ 10,524

26 – 40 yr 41 – 55 yr 56 – 70 yr

% 95% CI

n [ 3506 n [ 3689 n [ 3329

Hypertension 21.9 7.2 20.3 39.2 16.7 16.0 – 17.4

Diabetes 16.8 6.2 17.8 26.7 13.4 12.8 – 14.1

Obesity 24.2 18.3 28.2 25.9 23.2 22.4 – 24.0

CKD 6.7 3.0 5.3 12.0 5.1 4.7 – 5.5

eGFR < 60 2.7 0.5 1.3 6.6 1.6 1.4 – 1.8

Proteinuria 1.6 0.9 1.2 2.7 1.3 1.1 – 1.5

Hematuria 3.4 1.9 3.3 5.2 2.4 2.1 – 2.7

Corrected prevalence according to the Moroccan population distribution: 0.78 (41–55 years) and 0.29 (56–70 years).

Hypertension: con

rmed systolic blood pressure

140 mm Hg or diastolic blood pressure

90 mm Hg. Diabetes: glycemia

1.26 g/l. Obesity: body mass index

30 kg/m

. Proteinuria and hematuria: con

rmed

1 cross positive.

CI, confidence interval; CKD, chronic kidney disease; eGFR, estimated glomerular

c l i n i c a l i n v e s t i g a t i o n M Benghanem Gharbi et al.: MAREMAR study

In contrast to the overdiagnosis of CKD, the “ underdiag- nosis ” received much less attention. In our study, we found 130 subjects (1.2% of study population) with unclassi fi able CKD stage, that is, no proteinuria and eGFR > 60 ml/min/

1.73 m

, and having a low eGFR for their age (below the 3rd percentile for age and sex category). Young and middle-aged adults with an eGFR of 60 to 74 ml/min/1.73 m

but without proteinuria and whose life expectancy is probably substan- tially shorter than those with higher levels of eGFR are not meeting the criteria for CKD, whereas a large number of older adults with eGFR levels slightly below 60 ml/min/1.73 m

and no proteinuria, whose life expectancy is probably similar to those with higher levels of eGFR, are classi fi ed as CKD3A.

Risk strati fi cation in elderly patients should not be based on the sole eGFR cutpoints as for younger age groups and would bene fi t from fi ner categorization of the 30 to 59 ml/min/1.73 m

GFR.

A number of age and sex-speci fi c percentile lines of eGFR (most using MDRD formulas) or measured GFR were published.

It is of interest to observe, as in our study, that the 60 ml/min/1.73 m

value line hits the third to fifth percentile lines between 50 and 60 years of age in men and at a younger age in women. It seems that the physiological aging process of the kidney, along with a decrease in renal func- tion,

is rather comparable across populations of different races.

Although hematuria is mentioned in every edition of KDIGO guidelines, in almost all studies this parameter was not studied. The prevalence of con fi rmed hematuria was clearly more prevalent in women, only a minority of the subjects had the combination of hematuria and proteinuria.

We have not been able currently to de fi ne a renal or urological origin. Hematuria was found in 1.9% of women 25 to 40 years of age and increases up to 5.2% in those 56 to 70 years, rendering menstruation a limited source of hema- turia. The inclusion of hematuria in the screening program obviously results in an increase in the number of CKD1 and 2 subjects.

In the current study, the prevalence of hypertension was 16.7%, which is lower than was found (32.3%) in a previous Moroccan study.

In our study, blood pressure was deter- mined by an average of up to 3 measurements repeated after an interval of at least 7 days, while in the fi rst survey, an average of 2 measurements on the same day. This underscores again the importance of chronicity of increased blood pres- sure to eliminate incorrect classi fi cation.

One striking study result is the pronounced obesity in women, already present in the age group of 26 to 40 years.

Obesity in North Africa is linked to a greater availability of food, particularly from the West, and an increasingly seden- tary lifestyle in urban areas.

A woman with a low level of education is more likely to be obese. She, along with the general public, is not aware of the medical consequences of obesity. Rather, female fatness is viewed as a sign of social status, a cultural symbol of beauty, fertility, and prosperity.

Being thin is a sign of sickness or poverty.

In our study, an

overall increased eGFR in women in comparison to men was observed. A comparable signi fi cant increase in measured GFR (chromium-51 – labeled ethylenediamine tetraacetic acid) in obese women compared to matched control subjects was found in another study.

On the other hand, no arguments in favor of increased prevalence of CKD among obese individuals were found in the MAREMAR study.

MAREMAR demonstrates that the choice of arbitrary single threshold of eGFR for classifying CKD3 to 5 inevitably leads to overdiagnosis (false positives) of CKD in the elderly, particularly those without proteinuria, hematuria, or hyper- tension. It also leads to underdiagnosis (false negatives) of CKD in younger individuals with an eGFR above 60 ml/min/1.73 m

and below the third percentile of their age/sex category. The use of a third percentile eGFR level as cutoff based on age-/sex-speci fi c reference values of eGFR of the Moroccan population allowed the detection of these false positives and negatives. Lack of con fi rmation of proteinuria and “ chronicity ” of a decreased eGFR are the main reasons for the inflation of the CKD prevalence in published reports. The combination of population screening encompassing 4 different major health problems in the same screening procedure, using the correct methodologies and procedures, combined with a prevention/follow-up program results in a clinically/scientifically relevant program.

Clinical follow up over the next 8 years of subjects with CKD3A, 3B, and 4 and young subjects with a low eGFR for their age (below the third percentile for age and sex category) found in the context of the MAREMAR study has been installed.

MATERIALS AND METHODS

The MAREMAR cross-sectional study was based on a stratified random sample of the adult population of 2 Moroccan towns, El Jadida (industrial) and Khemisset (rural).

Sampling

Each town was divided into 5 areas served by 1 of its 5 health centers.

All inhabitants of the area ages 26 to 70 were categorized into 6 strata based on age (per 15 years) and sex using recent voting lists. Within each stratum, 167 effective participants and 100 reserves were randomly selected for each health care area. Sampling took into account a likely nonresponse rate of 20% to 40%.

Sample size calculation

In November 2004, the population between 26 and 70 years of age of the 2 towns was 113,962 inhabitants: El Jadida 67,309; Khemisset 46,553. Recruitment was planned over a period of 6 months based on the assumption that 8 subjects were screened per working day in each health center, resulting in a sample size of 10,000 participants (9% of eligible population). With this sample size of 10,000 par- ticipants, the 95% CI for the observed prevalence of CKD was expected to be <1% (supplementary data).

Recruitment

A trained nurse visited all selected subjects at home to explain in a

visual way the aims and methods of the program and benefits for the

participant. A tube for the collection of urine was given and the

subject was informed using the picture book, how to collect a first

morning mid-stream void urine sample (Supplementary Figure S2).

An appointment was made for a visit at the local health center for the baseline renal screening. If a selected subject could not be contacted during the first visit, the nurse returned. If this second visit turned out to be negative, the nurse would give the information, leave the urine tube, and an invitation via a proxy person. In case of an in- dividual not showing up, another subject was selected from the reserve list. Only people between 26 and 70 years of age with normal mental health were included. Pregnant women, cancer patients, and patients with paralysis were excluded. Women were asked for their menstruation period. If they were menstruating screening was postponed.

Data collection

Clinical investigations consisted of measurement of body weight, height, and blood pressure. Blood pressure was measured, at right upper arm, in sitting position (5 minutes’ rest) the right arm on a desk, using an automatic blood pressure monitor (Spengler M 20, Issoudun, France) taking 3 measurements at 15 seconds intervals, delivering an average of these 3. Hypertension was defined as systolic blood pressure $140 mm Hg and/or diastolic blood pressure >90 mm Hg. Fasting glycemia was measured enzymatically using a glucose oxidase method. Plasma creatinine was measured using an improved Jaffé method (isotope-dilution mass spectrometry) with a standard traceable to the serum creatinine reference system (National Institute of Standards and Technology).

The eGFR (ml/min/1.73 m

) was calculated with the most frequently used MDRD 4 variable formula: eGFR ¼ 175 (C)

(Age)

(0.742 if female) (1.212 if black).

The early morning mid-stream urine sample was used for semiquantitative dipstick analysis (Combiscreen 7 SYS þ; Alere Inc., Waltham, MA). Interpreted as follows: negative; þ:

mild; þþ or more: overt proteinuria. In case of proteinuria, hema- turia, or abnormal blood pressure, subjects were invited to visit the health center during the next 2 to 4 weeks to validate the observa- tions made before. When blood samples showed hemolysis, a second blood sample was taken. Chronicity of decreased eGFR was evalu- ated after 3, 6, and 12 months. All subjects with confirmed renal pathological findings (confirmed proteinuria, hematuria, and or eGFR below 60 ml/min/1.73 m

) were sent to the nephrologist of the local hospital for a further diagnostic work-up, individualized treatment, and follow-up.

Statistics

Descriptive analysis was performed using the statistical package IBM-SPSS (version 21; Armonk, NY). Continuous measures were summarized by mean SD and frequencies and categorical vari- ables by proportions for categorical variables; comparisons of continuous variables were done by Student t-test and analysis of variance and comparison of categorical variables were done by chi- square test. A 95% CI was calculated for the adjusted prevalence. For the calculation of the adjusted prevalence, the correction was based on the demographic structure of the general population of Morocco ages 26 to 70 years (according to census data 2004) as follows:

(i) ages 41 to 55 years: correction factor 0.78 (size of this population is only 78% of group ages 26 to 40); (ii) ages 56 to 70 years:

correction factor 0.29 (size of this population is only 29% of group ages 26 to 40).

DISCLOSURE

All the authors declared no competing interests.

ACKNOWLEDGMENTS

The study was funded in part by the Ministry of Health of Morocco, the World Health Organization, the International Society of Nephrology, and the Moroccan Society of Nephrology.

ETHICAL CONSIDERATIONS

The protocol of the MAREMAR study was approved by the Ethical Commission of the Faculty of Medicine of the University of Casablanca.

An informed consent letter was available in Arabic and French and signed by all participants. In case of illiteracy, a proxy signed.

SUPPLEMENTARY MATERIAL

Figure S1. Recruitment in the different health centers in the 2 cities.

Figure S2. Urine sampling —fi gure in the booklet given to the randomized subjects of how to perform a midstream urine.

Random selection of inhabitants of Khemisset.

The MAREMAR study team.

Supplementary material is linked to the online version of the paper at www.kidney-interntanional.org.

REFERENCES

1. Eckardt KU, Coresh J, Devuyst O, et al. Evolving importance of kidney disease: from subspecialty to global health burden. Lancet. 2013;382:

158 – 169.

2. Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group.

KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int. 2013;(suppl 3):1 – 150.

3. Luyckx VA, Naicker S, McKee M. Equity and economics of kidney disease in sub-Saharan Africa. Lancet. 2013;382:103 – 104.

4. Naicker S. Burden of end-stage renal disease in sub-Saharan Africa. Clin Nephrol. 2010;74(suppl 1):S13 – S16.

5. McCullough K, Sharma P, Ali T, et al. Measuring the population burden of chronic kidney disease: a systematic literature review of the estimated prevalence of impaired kidney function. Nephrol Dial Transplant. 2012;27:

1812 – 1821.

6. Brück K, Jager KJ, Dounousi E, et al. Methodology used in studies reporting chronic kidney disease prevalence: a systematic literature review. Nephrol Dial Transplant. 2015;30(suppl 4):iv6 – iv16.

7. Zhang QL, Rothenbacher D. Prevalence of chronic kidney disease in population-based studies: systematic review. BMC Public Health. 2008;8:117.

8. Anothaisintawee T, Rattanasiri S, Ingsathit A, et al. Prevalence of chronic kidney disease: a systematic review and meta-analysis. Clin Nephrol.

2009;71:244 – 254.

9. Zhang L, Wang F, Wang L, et al. Prevalence of chronic kidney disease in China: a cross-sectional survey. Lancet. 2012;379:815 – 822.

10. Hallan SI, Coresh J, Astor BC, et al. International comparison of the relationship of chronic kidney disease prevalence and ESRD risk. J Am Soc Nephrol. 2006;17:2275 – 2284.

11. Coresh J, Selvin E, Stevens LA, et al. Prevalence of chronic kidney disease in the United States. JAMA. 2007;298:2038 – 2047.

12. Wen CP, Cheng TY, Tsai MK, et al. All-cause mortality attributable to chronic kidney disease: a prospective cohort study based on 462 293 adults in Taiwan. Lancet. 2008;371:2173 – 2182.

13. Perkovic V, Cass A, Patel AA, et al., for the InterASIA Collaborative Group.

High prevalence of chronic kidney disease in Thailand. Kidney Int.

2008;73:473 – 479.

14. Hooi LS, Ong LM, Ahmad G, et al. A population-based study measuring the prevalence of chronic kidney disease among adults in West Malaysia.

Kidney Int. 2013;84:1034 – 1040.

15. Mills KT, Xu Y, Zhang W, et al. A systematic analysis of worldwide population-based data on the global burden of chronic kidney disease in 2010. Kidney Int. 2015;88:950 – 957.

16. Myers GL, Miller WG, Coresh J, et al., for the National Kidney Disease Education Program Laboratory Working Group. Recommendations for improving serum creatinine measurement: a report from the Laboratory Working Group of the National Kidney Disease Education Program. Clin Chem. 2006;52:5 – 18.

17. Levey AS, Coresh J, Greene T, et al., for the Chronic Kidney Disease Epidemiology Collaboration. Using standardized serum creatinine

c l i n i c a l i n v e s t i g a t i o n M Benghanem Gharbi et al.: MAREMAR study

values in the modi fi cation of diet in renal disease study equation for estimating glomerular fi ltration rate. Ann Intern Med. 2006;145:

247 – 254.

18. Qaseem A, Wilt TJ, Cooke M, Denberg TD. The paucity of evidence supporting screening for stages 1 – 3 CKD in asymptomatic patients with or without risk factors. Clin J Am Soc Nephrol. 2014;9:1993 – 1995.

19. Pugia MJ, Wallace JF, Lott JA, et al. Albuminuria and proteinuria in hospitalized patients as measured by quantitative and dipstick methods.

J Clin Lab Anal. 2001;15:295 – 300.

20. Glassock RJ. Is the presence of microalbuminuria a relevant marker of kidney disease? Curr Hypertens Rep. 2010;12:364 – 368.

21. Lin J, Knight EL, Hogan ML, Singh AK. A comparison of prediction equations for estimating glomerular fi ltration rate in adults without kidney disease. J Am Soc Nephrol. 2003;14:2573 – 2580.

22. Stevens LA, Coresh J, Greene T, Levey AS. Assessing kidney function – measured and estimated glomerular fi ltration rate. N Engl J Med.

2006;354:2473 – 2483.

23. Eriksen BO, Ingebretsen OC. In chronic kidney disease staging the use of the chronicity criterion affects prognosis and the rate of progression.

Kidney Int. 2007;72:1242 – 1248.

24. Inker LA, Tighiouart H, Aspelund T, et al. Lifetime risk of stage 3 – 5 CKD in a community-based sample in Iceland. Clin J Am Soc Nephrol. 2015;10:

1575 – 1584.

25. Eriksen BO, Ingebretsen OC. The progression of chronic kidney disease: a 10-year population-based study of the effects of gender and age. Kidney Int. 2006;69:375 – 382.

26. Glassock RJ. Con: thresholds to de fi ne chronic kidney disease should not be age dependent. Nephrol Dial Transplant. 2014;29:774 – 779.

27. Levey AS, Coresh J. Chronic kidney disease. Lancet. 2012;379:165 – 180.

28. Brantsma AH, Bakker SJ, Hillege HL, et al., for the PREVEND Study Group.

Cardiovascular and renal outcome in subjects with K/DOQI stage 1 – 3

chronic kidney disease: the importance of urinary albumin excretion.

Nephrol Dial Transplant. 2008;23:3851 – 3858.

29. O ’ Hare AM, Bertenthal D, Covinsky KE, et al. Mortality risk strati fi cation in chronic kidney disease: one size for all ages? J Am Soc Nephrol. 2006;17:

846 – 853.

30. O ’ Hare AM. Measures to de fi ne chronic kidney disease. JAMA. 2013;309:

1343.

31. de Jong PE, Gansevoort RT. Fact or fi ction of the epidemic of chronic kidney disease – let us not squabble about estimated GFR only, but also focus on albuminuria. Nephrol Dial Transplant. 2008;23:1092 – 1095.

32. Coresh J, Byrd-Holt D, Astor BC, et al. Chronic kidney disease awareness, prevalence, and trends among U.S. adults, 1999 to 2000. J Am Soc Nephrol. 2005;16:180 – 188.

33. Elseviers MM, Verpooten GA, De Broe ME, De Backer GG. Interpretation of creatinine clearance. Lancet. 1987;1:457.

34. Wetzels JF, Kiemeney LA, Swinkels DW, et al. Age- and gender-speci fi c reference values of estimated GFR in Caucasians: the Nijmegen Biomedical Study. Kidney Int. 2007;72:632 – 637.

35. Rule AD, Gussak HM, Pond GR, et al. Measured and estimated GFR in healthy potential kidney donors. Am J Kidney Dis. 2004;43:112 – 119.

36. Baba M, Shimbo T, Horio M, et al. Longitudinal study of the decline in renal function in healthy subjects. PLoS One. 2015;10:e0129036.

37. Zhou XJ, Rakheja D, Yu X, et al. The aging kidney. Kidney Int. 2008;74:710 – 720.

38. Tazi MA, Abir-Khalil S, Chaouki N, et al. Prevalence of the main cardiovascular risk factors in Morocco: results of a National Survey, 2000.

J Hypertens. 2003;21:897 – 903.

39. Rguibi M, Belahsen R. Fattening practices among Moroccan Sahraoui women. East Mediterr Health J. 2006;12:619 – 624.

40. Stokholm KH, Brøchner-Mortensen J, Hoilund-Carlsen PF. Increased glomerular fi ltration rate and adrenocortical function in obese women.

Int J Obes. 1980;4:57 – 63.