Ghana-Michigan Charter, C/O Office of the Provost, College of Health Sciences, University of Ghana, P. O. Box KB 52, Korle-Bu, Ghana
Department of Medicine and Therapeutics; National Diabetes Management and Research Centre, University of Ghana Medical School, P. O. BOX 4236, Korle-Bu, Ghana
Consultant Family Physician and CEO Mission Clinic, P. O. Box AN 10545, Accra North, West Africa, Ghana
Department of Health Behaviour and Health Education, University of Michigan School of Public Health, 1415 Washington Heights 3814 SPH I, Ann Arbor, Michigan, 48109, USA
Abstract
Background
The Metabolic syndrome (MS) which is a constellation of cardiometabolic risk factors including dyslipidaemia, hypertension, hyperglycaemia, central obesity, and endothelial dysfunction was hitherto relatively uncommon among Africans south of the Sahara. This study seeks to determine the prevalence of MS, its components and risk factors among a rural population in Ghana based on two popular international algorithms.
Methods
This was a cross-sectional survey of a rural population in Ghana conducted between November and December, 2007. Two hundred and twenty-eight (228) settler farmers, families and staff associated with the GOPDC Ltd, between the ages of 35 and 64 years, were randomly selected for the study; pregnant women were excluded. The prevalence of MS was estimated using the IDF and ATPIII criteria.
Results
The final subject pool included 102 males, and 104 females. The mean age of all subjects was 44.4 ± 6.9 years. The overall prevalence of MS by the IDF and ATPIII criteria were 35.9% and 15.0%, respectively, but there was an alarming female preponderance by both criteria {IDF: males = 15.7%, females =55.8%; ATPIII: males = 5.9%, females = 24.0%; sex differences p<0.001 for both criteria}. The most important determinants for IDF-defined MS were central obesity (55.3%), low High Density Lipoprotein (42.7%) and high Blood Pressure (39.5%).
Conclusion
The triad of central obesity, high blood pressure and low HDL were most responsible for the syndrome in this rural population.
Background
Metabolic syndrome (MS) which hitherto was relatively uncommon among Africans south of the Sahara, is increasingly becoming a public health concern in recent times
Several criteria have evolved in defining the syndrome. In 1998, the World Health Organisation (WHO) published the first working definition of MS with emphasis on insulin resistance. Subsequently, the European Group for the Study of Insulin Resistance (EGIR) proposed a modification to that of the WHO. The EGIR criterion still required evidence of insulin resistance but added greater focus on abdominal obesity. In 2001, the National Cholesterol Education Programme (NCEP) Adult Treatment Panel III (ATP III) released its definition for MS, de-emphasizing insulin resistance. Subsequently in 2003, the American Association of Clinical Endocrinologists (AACE) modified the NCEP ATP III criterion to refocus on insulin resistance as the primary cause of metabolic risk factors. In the wake of the varied algorithms for MS and consequential confusion in comparing epidemiological studies, the International Diabetic Federation (IDF) in 2005 provided a clinical algorithm for MS that attempts to accommodate the different diagnostic criteria and ethnic differences
High prevalence rates of over 40% have been documented by the IDF criterion in Portugal, Mexico, Urban China, and the United Arab Emirates. India and Brazil also recorded rates above 40% by the NCEP ATP III criterion. Low rates below 10% by both the NCEP ATP III and IDF criteria have been reported in Spain, Japan and Hong Kong
Studies on the MS in sub-Saharan Africa is scanty
Methods
Subjects
This was a cross-sectional survey of a rural population in Ghana conducted from November to December 2007. In arriving at the sample size, an assumed prevalence rate of 50% was used since there was no record of any study on MS in the country. At 95% confidence interval and the degree of accuracy set to 0.05, the desired sample size came to 384. Excluding pregnant women, hypertensives, diabetics, persons not willing to participate in the study and others who were on vacation or could not be traced during the period of the study, the sample size reduced to 228. The subjects included settler farmers, families and staff associated with the Ghana Oil Palm Development Company Limited (GOPDC Ltd), between the ages of 35 and 64 years. Subjects were randomly selected from the staff list in the accounts office and nominal roll in the Human Resource Unit of the Company. Randomisation was facilitated by coded tally cards. Data for 22 subjects out of the 228 were incomplete and discarded as a result of insufficient blood samples.
Measurements
Two professional nurses were trained on the structured interview guide for the survey. The interview guide was divided into three sections: socio-demographic factors, anthropometric measures and biochemistry.
The interview guide gathered information on demographic factors like age, gender and educational background; risk factors of chronic diseases such as smoking, alcohol intake, diet and physical exercise; prevalence of chronic diseases including hypertension, diabetes mellitus, and dyslipideamia.
The anthropometric measures taken included height, body weight measured in the upright position to the nearest 0.5cm and 0.1kg respectively. The waist circumference (WC) measurements were taken at the end of a normal expiration to the nearest 0.1cm, measuring at the midpoint between the subcostal plane and the supracristal plane.
Mercury sphygmomanometers were used to measure the blood pressure of each subject in the sitting position after 30 minutes of rest. Subjects were asked to refrain from smoking, or ingesting alcohol and caffeine containing products a day before the survey. Three readings each of systolic and diastolic blood pressures were recorded per subject with an interval of five minutes at the least and the mean was used for the data analysis.
Ethical review and ethics in human subjects research
Ethical approval for this study was given by the Ghana Health Service Ethics Review Committee. Reviewers from the West African College of Physicians have also reviewed the research protocol and granted permission for its execution in fulfilment of requirements towards the award of a Fellowship in Family Medicine. Administrative permission was granted by St. Dominic’s Hospital and the Ghana Oil Palm Development Company where the research was carried out.
All respondents voluntarily participated after the intent and the design of the study had been explained to them and signing informed consent forms prior to implementation of the study.
Biochemical analysis
Blood sampling was done on the mornings of six Saturdays in the months of November and December 2007 among subjects who had completed the first and second sections of the interview guide. Announcement was made through the community using a public address system the evening preceding blood sampling. Samples were obtained from antecubital veins using 10ml syringes after an overnight fast (10 – 16 hours). Samples for fasting plasma glucose were collected into sodium fluoride/K3EDTA bottles and that for fasting lipids were collected into vacutainer serum separated tubes.
Blood samples were immediately analysed for fasting blood glucose using glucometer (“onetouch ultra”) and subjects found to be having abnormal readings were referred to the hospital. Urine samples were also collected and immediately analysed using the 10 parameter test strip. Urine ketones, glucose and proteins were among the parameters under investigation.
The remaining samples were packaged into ice chests and transported to the biochemistry laboratory of St. Dominic’s Hospital (SDH), Akwatia. Samples for fasting glucose and lipid were centrifuged at 5000RPM for 2 minutes and the supernatant plasma and serum respectively collected into plane eppendorf bottles. These were frozen and stored at 4°C.
Samples were collected in duplicate and one set was analysed in SDH whiles the second set was transported in an ice chest to the research laboratory in Korle-Bu Teaching Hospital, Accra within 24 hours for repeat biochemical analysis.
Fasting plasma glucose (FPG) was determined using the enzymatic photometric test according to Barham and Trinder via the “smartlab auto-analyser”. Fasting lipid profile including triglycerides (TG), high density lipoprotein (HDL), low density lipoprotein (LDL), and total cholesterol (TC) were also estimated using the “smartlab auto-analyser”.
TG was measured by the colorimetric enzymatic test using glycerol-3-phosphate-oxidase with the aid of the auto-analyser. HDL was determined after the precipitation of chylomicrons, very low density lipoproteins and low density lipoproteins. Centrifugation left only HDL in the supernatant which was determined enzymatically. Total cholesterol was determined by the enzymatic photometric test. LDL was computed, thus:
Diagnostic criteria for metabolic syndrome (MS)
The diagnostic criteria for the MS are listed as follows:
International Diabetes Federation (IDF) criterion:
WC ≥ 94cm in men or ≥80cm in women plus two or more of the following:
a. Low HDL = HDL <40mg/dl in males or <50mg/dl in females, or specific treatment for this lipid abnormality
b. Hypertriglyceridaemia = TG≥150mg/dl, or specific treatment for this lipid abnormality
c. Hypertension = SBP ≥130mmHg or DBP≥85mmHg, or treatment for previously diagnosed hypertension
d. Dysglycaemia = FPG≥100mg/dl, or previously diagnosed type 2 diabetes
National Cholesterol Education Program Adult Treatment Panel (NCEP ATP III) criterion:
At least three of the following criteria:
a. Obesity = WC>102cm in men and 88cm in women
b. Hypertrigylceridaemia = TG≥150mg/dl
c. Low HDL = HDL<40mg/dl in men and <50mg/dl in women
d. Hypertension = BP≥130/85mmHg
e. Dysglycaemia = FPG ≥110mg/dl
Statistical analysis
Data was analysed using the Statistical Package for Social Sciences, version 15.0 software (SPSS Inc., Chicago, Illinois, USA). Continuous variables were expressed as means and standard deviations and discrete variables were presented as proportions. The prevalence of the metabolic syndrome (MS) among males and females was determined. Age-specific prevalence rates of MS and its components were also determined.
The chi-squares test was performed to test for differences in proportions of categorical variables between two or more groups. In 2*2 tables, the Fisher exact test (2-tailed) replaced the chi-squared test if the assumptions underlying chi-squared were violated, such as situations of small sample size and where the expected frequency is less than 5 in any of the cells.
Multinomial logistic regression analysis was used to determine the type and degree of association between MS and its socio-demographic and behavioural risk factors. The result is presented as odds ratios (ORs) together with their 95% confidence interval (95%CI). The trend in ORs across MS risk factors was evaluated using the likelihood ratio test.
The agreement between IDF and NCEP ATP III criteria of MS was determined by the kappa statistics (k). The level of agreement is considered poor with k≤0.20, fair with k=0.21 to 0.40, moderate with k=0.41 to 0.60, substantial with k=0.61 to 0.80, and very good with k>0.80
Results
Data on 206 subjects, 102 (49.5%) males and 104 (50.5%) females were analysed. The socio-demographic and behavioural characteristics of the study population by sex are provided in Table
Active office workers (Junior staff) were the most prevalent, 55.3% among the study subjects and there was a female preponderance (67.3%). Smoking was rare, at 9.2%, and alcohol consumption was 55.1% prevalent overall with a high male prevalence (67.3%). More than half (57.8%) of the study subjects had normal weight; for males this was true for 70.6%. The obese constituted less than a fifth of the population, and 28.4% of females.
The basic characteristic of the study subjects by sex are shown in Table
Table
Central obesity (55.3%) was the most prevalent determinant of MS by the IDF criterion followed by low HDL (42.7%) and then high BP (39.5%). The NCEP ATP III criterion showed HDL (42.7%) as the most prevalent followed by high BP (39.5%) and central obesity (31.1%). The sex difference in prevalence rates among the determinants was only significant for central obesity and low HDL by both criteria (p<0.001, p= 0.005 respectively by both criteria). In both the IDF and NCEP ATP III, central obesity in women and high BP in men were the most common determinants and the triad of central obesity, low HDL and high BP constituted the most widespread combination of metabolic abnormalities determining MS.
Table
The predictive odds of MS (IDF criterion) with variations in socio-demographic and behavioural risk factors are shown in Table
Discussion
This study was conducted among staff, relatives and settler farmers associated with the GOPDC Ltd and thus the findings cannot be generalized to all rural communities in Ghana. In spite of this, these findings are very important to the documentation of metabolic syndrome in modern Ghana.
The prevalence of MS was alarmingly high {35.9% (IDF) and 15.0% (NCEP ATP III)} in this rural community of mainly farmers and small scale industrialist (oil palm extraction). Earlier studies in Cameroon
The IDF criterion yielded a higher prevalence rate (2.4 times) than the NCEP ATP III and this may be due to the lower cut-off points for WC and FPG. Similar trends have been observed in other studies
In spite of the differences observed in the prevalence rates by the IDF and NCEP ATP III criteria, the agreement between the two algorithms was moderate, 0.45 as alluded to by Choi and co-workers
There was a significant sex difference in the prevalence of MS in this population. The prevalence was approximately four times higher among females by both algorithms and this is in consonance with other studies with similar, though relatively less dramatic, sex differentials
The triad of low HDL, high BP, and central obesity were largely responsible for MS in this community. This combination was also observed in Nigeria
The prevalence of MS increased with age
Conclusion
In conclusion, the prevalence of an MS diagnosis was alarmingly high among this population of women in rural Ghana, despite high physical workloads. Consistent with experience elsewhere, the IDF algorithm identified twice as many subjects as having MS, compared to the NCEP ATP III algorithm. The triad of central obesity, high blood pressure and low HDL were most responsible for the syndrome in this rural population. Younger age, male sex, and normal weight were protective against MS. This study needs to be extended to other parts of the country to ascertain the national prevalence rate vis-à-vis, its rural–urban distribution. There is the additional need to determine whether these diagnostic standards are indeed associated with poor long-term clinical outcomes as they are in other parts of the world.
Abbreviations
MS: Metabolic syndrome; WHO: World health organisation; EGIR: European group for the study of insulin resistance; NCEP: National cholesterol education programme; ATP III: Adult treatment panel III; AACE: American association of clinical endocrinologists; IDF: International diabetic federation; GOPDC: Ghana oil palm development company limited; WC: Waist circumference; SDH: St. Dominic’s hospital; FPG: Fasting plasma glucose; TG: Triglycerides; HDL: High density lipoprotein; LDL: Low density lipoprotein.
Competing interests
No competing interests were declared by the authors.
Authors’ contributions
MKG conceptualised and executed the research and was the leading author of the manuscript. AGBA was part of the conceptualisation process and was instrumental in the analysis of the laboratory specimen and writing of the manuscript. DM also played a major part in conceptualising the research and was actively involved in the entire process from data collection through to manuscript production. RCS played a major part in data analysis and manuscript production. All authors read and approved the final manuscript.
Acknowledgements
We are indebted to Mr. Patrick Yankey, Mr. Nkumsah, Management and staff of GOPDC Ltd; Mr. Evans Chinebuah, Mrs. Rebecca Steele-Dadzie and staff of the national diabetes management and research centre for their valuable assistance. We wish to thank Dr. Mercy Dawson and the eye team, Mr. Shaibu and the laboratory team as well as management and staff of St. Dominic’s Hospital, Akwatia, Ghana for their support in the conduct of this study. We are very much indebted to Ms. Fidelia Dake in putting this final piece together.
Pre-publication history
The pre-publication history for this paper can be accessed here: