Lipid Profile of Type 2 Diabetic Patients at a Tertiary Hospital in Tanzania : Cross Sectional Study

Nyasatu G. Chamba1,2,Elichilia R Shao1,2,3,4,5,∗,Tolbert Sonda1,3and Isaack A. Lyarru1,2 1Department of Internal medicine Kilimanjaro Christian Medical University College, Tumaini University, Makumira PO BOX 2240, Moshi Tanzania 2Department of Internal Medicine Kilimanjaro Christian Medical Centre, PO BOX 3010 Moshi Tanzania 3Kilimanjaro Clinical Research Institute, PO BOX 2236, Moshi Tanzania 4Better Human Health Foundation, PO BOX 1348, Moshi Tanzania 5Imagedoctors Organization, PO BOX 16341, Arusha Tanzania


Background
Patients with diabetes mellitus are at high risk of cardiovascular events because of abnormal lipid status.Dyslipidemia is common in diabetes mellitus and is associated with cardiovascular complications.Early diagnosis and treatment is the main cornerstone in the prevention of its multiple complications.There is scarcity of data on the magnitude and risk factors associated with dyslipidemia among diabetic patients in the Northern zone of Tanzania.

Objective
The aim of the study was to determine the prevalence of abnormal lipid profile levels among patients attending the diabetes clinic at Kilimanjaro Christian Medical Centre, Moshi Tanzania.

Methodology
A cross-sectional study was conducted from October 2012 to March 2013 involving 119 diabetic patients, who were 18 years and above from the diabetes clinic at Kilimanjaro Christian Medical Centre, Moshi.A structured questionnaire was administered to evaluate the socio-demographic and clinical characteristics.Standard procedures were followed for measuring blood pressure, body mass index and waist circumference.Fasting blood samples were taken to measure lipid profiles and glycosylated hemoglobin.Data analysis was done by Statistical Package for Social Science version 21 statistical software and chi-squared test, and multiple logistic regression were used for data analysis and p-value of <0.05 was considered to be statistically significant.

Results
The prevalence of dyslipidemia among diabetic patients was 83%.Among the abnormal lipid profile levels, elevated Low Density Lipoprotein Cholesterol constituted the highest single abnormality having 67.23%.Dyslipidemia was mostly seen in the females (88.9%), advanced age of 50 years and above (86.7%),poor glycemic control (glycosylated hemoglobin >7%) was 60.7% and BMI >25kg/m2 was 91.7%.The multivariate logistic regression showed that BMI was the sole determinant for the development of dyslipidemia.

Conclusions
The prevalence of dyslipidemia is high.The sole determinant for diabetic dyslipidemia among our study population was increase in the body mass index.There is an urgent need for effective strategies for primary prevention of obesity, diagnosis and treatment of dyslipidemia among diabetic patients.

Background
Diabetes Mellitus (DM) is a metabolic disorder due to hyperglycemia, thus having disturbances in carbohydrates, lipid and protein metabolism which results from defects in insulin secretion, insulin action and or both [1].Globally, diabetes mellitus has been a major health care problem.In sub-Saharan Africa (SSA), diabetes represents 4.2% of the global population [2] it is estimated that by 2025, the number will increase by 80%, with a higher prevalence being in the urban areas [2,3].Dyslipidemia is a disorder which arises as a result of abnormalities in the plasma lipoproteins.The lipid abnormalities in diabetes include quantitative changes which occur due to an increase of glucose for very low density lipoprotein (VLDL) synthesis and decrease in lipoprotein lipase activity leading to decrease of VLDL from peripheral circulation, increase in low density lipoprotein-C (LDL-C) levels and decrease in high density lipoprotein C (HDL-C) levels due to increase in hepatic activity decrease in VLDL clearance.Qualitative changes consists of increase of triglyceride (TG), LDL-C and HDL-C, non-enzymatic glycation of LDL and non-enzymatic glycation of high density lipoprotein (HDL) [4].Due to the abnormalities in lipoproteins, diabetes mellitus is associated with cardiovascular and cerebrovascular morbidity and mortality worldwide [5].Dyslipidemia affects approximately 70% to 97% of people with diabetes [6].World Health Organization (WHO) in 2002 reported that dyslipidemia accounted for 18% of ischemic heart disease, 56% of stroke and over 4million deaths per year globally [7].The prevalence of dyslipidemia is terrifyingly high within the African continent.The trend has been seen to increase, in Nigeria, the prevalence ranged from 82.6% to 90.7% from 2008 to 2011 respectively [8,9,10].In South Africa the prevalence is over 90% [11].In Tanzania, the prevalence of diabetic dyslipidemia was 95% in 2007 [12].This is mainly due to the adopted western diet, sedentary lifestyle as well as physical inactivity resulting to obesity [13].By determining the prevalence of abnormal lipid profile levels among the diabetic patients, it will provide the need to aggressively manage dyslipidemia among diabetic patients.

Patients and Methods
A cross-sectional study was carried out at KCMC, Moshi Tanzania from October 2012 to March 2013.The study participants included all DM patients who were consecutively recruited, aged 18 years and above receiving treatment at the KCMC diabetic clinic during the study period.Patients on lipid lowering agents, pregnant women, HIV/AIDS on treatment and those with renal failure were excluded.Relevant medical history was completed and physical examination was conducted on all subjects who had an overnight fast for at least 8 hours.Ethical clearance was secured from Kilimanjaro Christian Medical University College (KCMUCo) Ethical Committee and signed consent to take part in the research and to publish results was obtained from the participants.Anthropometric measurements, including weight, height and waist circumference were measured with the subjects wearing light clothing and no shoes.Body Mass Index (kg/m 2 ) was calculated as weight (kilograms) divided by squared height in (meters).The study populations were classified as underweight BMI <18.5 kg/m2, normal BMI 18.5 ≥;BMI < 25 kg/m 2 and overweight BMI≥;25 kg/m 2 and obese BMI ≥;30 kg/m2 [14].Central obesity was considered as measurements above 102cm and 88cm in men and women respectively.Blood pressure was measured twice for each patient after at least 5 minutes of rest, by the use of a standardized mercury sphygmomanometer.Glycosylated hemoglobin test was done on the spot using the A1CNOW PLUS kits from the Bayer manufacturer and the results were ready in 5 minutes.Poor glycemic control was defined as glycosylated hemoglobin as >7%.After an overnight fast, approximately 4mls of venous blood was obtained from each patient for lipid profile (TC, TG, LDL-C and HDL-C) analysis.The samples were analyzed using the chemical analyzer COBAS INTEGRA 400 Plus serial NO 397672 in the main Kilimanjaro Christian Medical Centre (KCMC) laboratory.Dyslipidemia was taken as derangement in any of the lipid components: TC> 5.2 mmol/l, LDL-C >2.6 mmol/l, HDL-C < 1.1 mmol/l for males, < 1.38 mmol/l for females and TG > 1.7 mmol/l [15].Data analysis was done using Statistical Package of Social Sciences (SPSS) version16.Analysis of variance (ANOVA), t-test and chi-square test were used.Regression analysis was used to determine the predictors of dyslipidemia.A 95% confidence was used for the determination of significance of probabilities, and p-value <0.05 was regarded as statistically significant.Patients had to sign the consent form before enrolled in the study.Those found with dyslipidemia were started on lipid lowering agents and advised on the lifestyle changes.Original data and supporting material are available upon request from the publishers.
Diabetic dyslipidemia was found in (94.1%) patients.The pattern of lipid abnormalities according to sex, age, BMI and central obesity are presented in (Table 2).High TG, high LDL-C, high TC and low HDL-C exhibited an increasing trend in the proportion of patients with dyslipidemia by the BMI and the differences were statistically significant (p<0.05).A similar trend was observed in the patients with central obesity compared to those without.The difference was statistically significant for high TC and low HDL-C and the overall dyslipidemia (p<0.05).
The following risk factors namely female sex, age above 50years, BMI (overweight and obese), poor glycemic control, central obesity and physical inactivity were associated with diabetic dyslipidemia among diabetic patients attending the diabetes clinic, the p-values were statistically significant (Table 3).Other variables namely duration of diabetes mellitus, the type of diabetes mellitus, smoking habits and hypertension were not statistically significant in the association of dyslipidemia.By multivariate logistic regression analysis, significant predictor for dyslipidemia among the diabetic patients was overweight and obesity (BMI >25kg/m 2 ), with p-value of 0.040, OR (95%CI) 0.2 (0.1-0.9).

Discussion
The overall prevalence of diabetic dyslipidemia in this study was found to be 83.0%.The prevalence is similarly high compared to the study done by Chattanda in Dar es Salaam, Tanzania where the prevalence was found to be 95% [12].Though for the past five years, the prevalence of diabetic dyslipidemia seems to be gradually declining, it is still quite high.The apparent decline can be explained by the interventions in which the Ministry   [8,9,10] and in South Africa [11].Other areas such as Tehran and USA, similarly high prevalence of dyslipidemia was also observed 88.9% and 97% respectively [17,18].The fairly high prevalence in this study may be attributed to the current trend towards urbanization and adoption to the western diet and lifestyle which was also reported by Njelekela et al [13].This study showed a high prevalence of dyslipidemia with the commonest lipid abnormality being elevated LDL-C (64%) followed by low HDL-C (56%).Similar findings were also noted in the Third US National Health and Nutritional Examination Survey and the Behavioral Risk factors Surveillance System where majority DM patients had LDL-C of 58% [19] in Gaborone [20] and in Ghana [21].Reasoning to this is that within the African ethnicity, African-Americans, it has been observed that the elevated LDL-C levels are a common subclass of dyslipidemia compared with other subclasses [16].Different observations from other studies where total cholesterol to be the common subclass in dyslipidemia.In Dar es Salaam, Chattanda observed that among the subclasses of dyslipidemia, 95% had high triglycerides [12].In Kenya, over 70% of the study participants had high total cholesterol levels [22] as well as in Jordan [23].The observed differences in the subclasses of dyslipidemia could be explained by the difference in ethnicity where Indians who tend to have elevated levels of total cholesterol, presented in the diabetic clinics.Majority of Indians reside in the urban regions within the big cities in Africa.In this case, the different cut off levels in classifying dyslipidemia could have also contributed to the difference.The high frequency of LDL-C among the female patients over 50 years, was over 90% in this study.It can be explained by the fact that over half of the women involved in our study were in the post-menopausal years who tend to lose the protective effect of estrogen on lipid metabolism.Thus this leads to higher prevalence of LDL-C in women compared to men, also mostly observed within the black race.On the other hand, Chattanda did not find any difference between males and females [12]; this could have been due to similar lifestyles in the study population and the racial differences.This is comparable to a study done in Pakistan, where they found that females above 60 years were more likely to have dyslipidemia [24].
The increase in BMI (≥;25.0kg/m 2 ) was an independent determinant of dyslipidemia among the diabetic patients in this study, which was strongly associated with the prevalence of dyslipidemia.Majority of the patients also had central obesity over 90%.Similarly in Gaborone [20] and South-East Nigeria [10] showed that overweight and obesity was strongly associated with elevated plasma lipid levels.A report in the National Health and Nutrition Examination Survey, 1999-2004, showed that the prevalence of dyslipidemia substantially increases with increased BMI [14].Despite the gender differences within our study population, it was clearly seen that an increase in BMI among diabetic patients is the strongest factor in the development of dyslipidemia.Physical inactivity mostly observed among the elderly patients due to arthritis could be another contributory factor of overweight and obesity as well.In contrast to the study done in Dar es Salaam, where poor glycemic control was the sole determinant leading to the development of dyslipidemia [12].The difference could be explained by the fact that most of the diabetic patients were on medications, mostly oral hypoglycemic drugs, majority having a good compliance henceforth, such treatment reduces the blood glucose level and most probably alters the overall diabetic dyslipidemia pathophysiology.

Conclusion
Increased BMI was the strongest determinant as it was found to be the sole associated risk factor on the high prevalence of dyslipidemia which is over 83.0% and the commonest dyslipidemia was elevated LDL-C.

Table 2 :
The pattern of lipid abnormalities in sex, age, BMI and central obesity among type II diabetic patients attending KCMC diabetes clinic from October 2012 to March 2013,(n= 119)

Table 3 :
Dyslipidemia associated risk factors among type II diabetic patients attendingKCMC diabetes clinic from October 2012 to March 2013 (n=119)