Study setting and design
Data collection for this cross-sectional survey took place between January to June 2018, in Garoua and Figuil, two cities of the North Region of Cameroon, with predominantly peulh population. Garoua, the administrative capital of both Benoue division and North region, has a population of approximately 568,760 inhabitants distributed across two health districts, namely Garoua I and II, each with 6 health areas. The population of Garoua I has been estimated at 272,461 inhabitants from 54,492 households, while that of Garoua II has been estimated at 296,299 inhabitants across 59,260 households. Garoua adult population comprises students, traders, civil servants, private and parastatal companies’ workers mainly from the cotton development company (SODECOTON®). Figuil is the capital of eponym subdivision in the Mayo Louti division. It has an estimated population of 123,517 inhabitants distributed in the Figuil health district across 11 health areas and 24,703 household. The adult population comprises essentially private sector workers from two companies: CHAUX ROCA® is specialised in the production of marble, calcium carbonate, slake lime and gravel of granite while CIMENCAM® produces cement, aggregate and concrete. This study was approved by the Cameroon National Ethics Committee, and all participants provided a written informed consent before enrolment.
The sample size was estimated assuming a 10% prevalence (P) of CKD among urban adults in Cameroon based on the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation , a precision (I) of 2%, a correction factor (K) for the cluster effect of 2, and a confidence interval of 1.96. Using these assumptions, a minimum of 432 subjects (N) were required, by applying the following formula N= [(Zα /2)2 PQ/ I2] x K. We used a multi-level cluster sampling approach whereby the health area was the first level and the household the second level. The number of participants to be recruited was equally distributed across health districts, giving a ratio of 144 participants per district. This number was then divided by the number of health areas per health district, to give a ratio of 24 participants per health area of the Garoua health district and 14 participants per health area of the Figuil health district. We determined the sampling interval by dividing the number of household in the health area by the number of participants to be recruited in that health area, assuming a maximum of 2 adults per household. The first household to be included and selection approach mirrored those used during national immunization days. Potential adult participants were sensitized for the survey through their district council, community leaders, posters, leaflets and words of mouth. For household included in the survey, all adults were further explained the purpose of the study and received a health promotion campaign on communicable and non-communicable diseases. Thereafter, a maximum of two adults (20 years and above) who have been residing in the household for at least three month were randomly selected from the household to take part in the survey. The random selection was done by blindly picking two names from a small bag. When there was only one adult or in case of non-response, the study team moved to the next household to repeat the procedure and recruit the remaining participants. Were excluded from the survey, people with serious mental or physical impairment (limb amputation or paralysis in particular), pregnant women or lactating mother and people with simultaneous leucocyturia and urine nitrites.
Final year undergraduate medical student collected data between 8 and 12 am during household surveys and only from participants who provided a written informed consent. Biological samples including urine and blood specimens were collected from 8 and 9 a.m. During face-to-face interview with participants, the scope of date collected included socio-demographic information (age, gender, education and occupation), and clinical information including personal history of diagnosed conditions (hypertension, diabetes, dyslipidaemia, gout and infectious diseases such as HIV, hepatitis B and C), lifestyles (alcohol consumption and smoking), use of nephrotoxins (herbal and street medicines), anthropometric (weight, height and waist girth) and blood pressure measurements. Blood pressure (BP) measurement followed to the World Health Organization (WHO) guidelines , and used an automated device (OMRON HEM705CP, Omron Matsusaka Co, Matsusaka City, Mie-Ken, Japan). BP was measured on both arms, while the participant was comfortably sitted, had been at rest for 30 minutes or more, and had consumed no tea, coffee or smoked cigarette within the preceding hour. The standard 23 x 12 cm cuff or larger size cuff for obese individuals was used. The average of three consecutive BP measurements from the arm with the higher values was used in all analyses.
A 50 ml clean container was used to collect the mid-stream second morning urine for dipstick, creatinine and albumin tests. After an overnight fast of 8h or more, a 3 ml of whole blood was also collected from the antecubital vein for serum creatinine, uric acid, lipid profile (including total and high density lipoprotein (HDL) cholesterol, triglycerides) and fasting glycemia. Fasting glycemia and dipstick urine tests were done at the site of sample collection. The remaining sample was transported in ice to the Garoua Regional Hospital’s laboratory for further processing and analyses. The CombiScreen 7SL PLUS 7 test strips (Analyticon Biotechnologies AG, D-35104 Lichentenfeis, Germany) was used for urine dipstick tests. Fasting glycemia was acquired using the One Touch Ultra® easy reader® (LifeScan Europe, Cilag GmbH International, Zug, Switzerland). Serum and urinary creatinine measurements were based on the kinetic modification of the Jaffé reaction with Human visual spectrophotometer (Human Gesellschaft, Biochemica und Diagnostica mbH, Wiesbaden, Germany) implemented on a Beckman creatinine analyzer (Beckman CX systems instruments, Anaheim, CA, USA). Urinary albumin was measured using pyrogallol red-molybdate complex with Teco diagnostics tests (Teco Diagnostics, Anaheim, CA, USA). Lipid profile [serum Triglycerides (TG), total (TC) and HDL cholesterol (HDL-C)] measurement used the enzymatic colorimetric methods, while low-density lipoprotein cholesterol (LDL-C) was calculated from the Friedwald’s equation as LDL-C=TC – (HDL-C+TG/5) [15-17].
All participants with a positive dipstick [protein (≥ traces), blood, leucocytes], and/or a fasting glycemia of at least 126 mg/dl (for those without prior diabetes), were offered confirmation tests two weeks later. For participant with positive dipstick on repeated test and/or estimated glomerular filtration rate (eGFR) < 60 ml/min/1.73 m2 using CKD-EPI formula, the chronicity of the condition was confirmed 3 months later via repeated tests. For persistent proteinuria (≥ traces) 3 months later, a urinary albumin/creatinine ratio (ACR) was performed.
Definitions and calculations
We calculated the body mass index (BMI, kg/m2) as weight (kg)/height (m)*height (m), and ranked participants as normal weight for 20≤BMI<25 kg/m2, overweight for 25≤BMI<30 kg/m2 or obese for BMI≥ 30 kg/m2. Hypertension was diagnosed in the presence of systolic (SBP) ≥140 mmHg and/or a diastolic blood pressure (DBP) ≥90 mmHg on two consecutive occasions two weeks apart, or ongoing use of BP lowering medications. The 24-hour albuminuria was estimated from Albumin/Creatinine ratio (mg/g). Estimated glomerular ﬁltration rate (eGFR, mL/min) was based on the CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equation . Serum creatinine from Jaffe reaction (SCrJaffe) was converted to standardized serum creatinine (SCrStandardized) to be used in CKD-EPI formula, via the formula SCrStandardized = 0.95*SCrJaffe – 0.10 . CKD was defined by the persistence after 3 months of albuminuria (ACR ≥ 30 mg/g) and/or low eGFR (< 60 ml/min/1.73 m2) following the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines . Thereafter, CKD was classified into GFR and albuminuria categories. GFR categories of CKD included: G1 (eGFR ≥ 90); G2 (eGFR: 60 – 89); G3a (eGFR: 45 – 59); G3b (eGFR: 30 – 44); G4 (eGFR: 15 – 29) and G5 (eGFR < 15). Albuminuria categories of CKD were: A1 (<30 mg/g); A2 (30 – 300 mg/g) and A3 (>300 mg/g). Diabetes mellitus was diagnosed in the presence of fasting glycaemia ≥ 126 mg/dl on two consecutive occasions, or use of blood glucose control agents. Hyperuricemia was defined as serum uric acid level ≥ 7.0 mg/dl or use of uric acid lowering drugs. Dyslipidemia was defined by total cholesterol (≥160 mg/dl) and/or triglycerides (≥150 mg/dl) and/or HDL cholesterol (<35 mg/dl) and/or LDL cholesterol (≥130 mg/dl) or use of lipids lowering agents.
Data were analysed using the SAS/STAT v9.4 software for Windows® (SAS Institute Inc., Cary, NC, USA). The survey analysis procedures (‘proc surveymeans’, ‘proc surveyreg’ and ‘proc surveylogistic’) were applied to account for the multilevel sampling design of the sample collection and sampling weights. Results are reported as means, counts and percentages and the accompanying 95% confidence intervals. The Taylor series linearization method was used to estimate the sampling error. To compare qualitative variables across groups, we used the Rao-Scott design-adjusted chi-square test. This test accounts for the sample design and allows inference to be made for the study population. Age and sex adjusted logistic regression models were used to investigate the predictors of CKD. A p-value <0.05 was used to indicate statistically significant results.