Socio-economic status, urbanization, and cardiometabolic risk factors among middle-aged adults in Tanzania.

OBJECTIVE: This study was conducted to examine the associations between socioeconomic status, urbanization, and cardio-metabolic risk factors in Tanzania. METHODS: Participants were 209 adults (45-66 years) in Dar es Salaam, Tanzania. A structured questionnaire was used to evaluate socioeconomic status and behavioral characteristics, including income, education, occupation, residence (urban, rural, mixed), dietary factors, and smoking. Blood samples were collected and analyzed to measure lipid profiles and fasting glucose levels. Cardiometabolic risk factors were defined using World Health Organization criteria. RESULTS: Urban residence and higher socioeconomic status were associated with decreased intake of traditional staple food (ugali), and increased consumption of meat products and beverages. Higher socioeconomic status was associated with a significant 3.5-kg/m2 higher BMI (p = 0.0001) and 8 cm higher waist circumference (p < 0.001), and a three-fold increase in the risk of obesity. Urban residence was associated with poorer lipid profile, including significantly higher total cholesterol, increased LDL cholesterol, but lower triglycerides, compared to rural residence. The prevalence of metabolic syndrome was high (38%), and was associated with increased socioeconomic status. CONCLUSION: Urban residence and higher socioeconomic status were important correlates of cardiometabolic risk factors, including obesity and poorer lipid profile. Primary prevention and health screening strategies are needed to target cardiometabolic risk factors in urban areas, to reduce the burden of cardiovascular disease in Tanzania.

Dietary habits, plasma polyunsaturated fatty acids and selected coronary disease risk factors in Tanzania.

OBJECTIVES: To compare dietary habits and percent compositions of omega (omega)-3 fatty acid (FA) in plasma phospholipids and to examine if there is any association between fish intake and selected coronary disease risk (CHD) factors in this population. DESIGN: Cross sectional population study. SETTING: Three areas in Tanzania: urban Dar es Salaam (D), rural Handeni (H) and pastoralist population of Maasai in Moduli (Mo). SUBJECTS: One hundred and five participants (Dar 36, Handeni 37 and Monduli 32) aged 47-57 years. MAIN OUTCOME MEASURES: Fatty acids composition in plasma phospholipids. RESULTS: The frequency of intake of fish, meat, coconut milk and fruits was higher in Dar than in Handeni or Monduli (all p< 0.0001). The Maasai from Monduli had the highest percentage consumption of milk in a week (p<0.0001) and lowest intake of fish (p<0.0001). Participants from Handeni had the highest prevalence of consumption of green vegetables (p<0.0001). Percent compositions of arachidonic acid (AA; C20: 4omega-6) and docosahexaenoic acid (DHA; C22: 6omega-3) in plasma phospholipids were the lowest in Monduli (p<0.0001). Selected coronary disease risk factors were higher in the urban area Dar. The frequency of intake of fish per week correlated negatively with total cholesterol (TC) and LDL-C and HBA1c percent but not with blood pressure. The percent composition of omega-3 FA in plasma phospholipids was positively correlated with the frequency of intake of fish a week. CONCLUSION: Our results indicate that, there are significant differences in dietary patterns among the three study areas, and that the intake of fish is inversely associated with selected risk factors for coronary heart disease.

Determinants of hyperleptinaemia in an African population.

OBJECTIVE: To examine the determinants for elevated plasma leptin concentration in normal weight (NW), obese (OB), and morbidly obese (MO) individuals in Tanzania. DESIGN: Cross-sectional epidemiological study, the CARDIAC study. SETTING: Three areas in Tanzania; Dar es Salaam, urban (U), Handeni, rural (R) and Monduli, pastoralists (P), in August 1998. SUBJECTS: Five hundred and forty five participants from a random sample of 600 people aged 46-58 years. MAIN OUTCOME MEASURES: Plasma leptin concentrations, height, weight, body mass index (BMI), lipid profiles, haemoglobin A1c (HBA1c), and blood pressure (BP). RESULTS: Plasma leptin concentrations were higher in women than in men (women; 16.0 ng/mL, men; 3.1 ng/mL; p<0.0001). Women showed a higher mean body mass index (BMI), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein cholesterol (HDL-C) than men. In both genders, plasma leptin concentration, total cholesterol (TC), low density lipoprotein cholesterol (LDL-C), triglycerides (TG), systolic BP (SBP) and diastolic BP (DBP) were significantly higher in OB than in NW participants. MO women had significantly higher leptin concentration, SBP and DBP compared with the other two groups. In NW men, log leptin concentrations showed a direct correlation with weight, BMI, HBA1c, TC, LDL-C, TG, SBP and DBP (all p<0.0001 except TG; p<0.001), while among NW women and OB men, weight and BMI correlated positively with log leptin (all p<0.05). OB women observed a positive correlation between log leptin and weight, BMI and LDL-C. Regression analysis indicated that among NW subjects, gender, BMI and TC explained 53.9% of the variation in log leptin. In OB subjects, gender, BMI and LDL-C explained 51.7% of the variability in leptin levels. No relationship was found between log leptin and CVD risk factors among MO subjects. CONCLUSION: The most important determinants for hyperleptinaemia in NW participants were gender, BMI, TC, while in addition to these LDL-C, was an important determinant of leptin concentration in OB individuals. In MO women, the high leptin concentrations did not reflect the amount of adipose stores.

Obesity and lipid profiles in middle aged men and women in Tanzania.

OBJECTIVE: To examine the relationship between obesity and lipid profiles and to compare the effects of total obesity and central adiposity on lipids in three locations in Tanzania. DESIGN: Cross-sectional epidemiological study. SETTING: Three areas in Tanzania: Dar es Salaam (urban), Handeni (rural) and Monduli (pastoralists), in August 1998. SUBJECTS: Five hundred and forty five men and women from a random sample of 600 people aged 46-58 years. MAIN OUTCOME MEASURES: Mean BMI, waist circumference, WHR, TC, HDL-C, LDL-C, TG and LDL/HDL ratio. Prevalence rates of overweight,obesity, central obesity and dyslipidaemia. RESULTS: As compared to men, women had higher BMI (24.7 versus 22.5 kg/m2, p<0.0001), waist circumference (92.4 versus 89.1 cm, p<0.05), TC (4.9 versus 4.2 mmol/L, p<0.0001) and LDL-C (3.3 versus 2.6 mmol/L, p<0.0001). The urban population demonstrated higher levels of lipid factors than the rural population (TC, men 4.8 mmol/L; women 5.3 mmol/L, p<0.0001; TG, men 3.6 mmol/L; women 3.7 mmol/L, p<0.0001, LDL-C, men 2.8 mmol/L, p<0.0001). BMI and waist circumference correlated positively with serum TC, TG, and LDL-C in both genders. Stepwise regression analysis showed that BMI predicted triglyceride concentration in men (p<0.05) and women (p<0.0001). Waist circumference predicted levels of TC in women only (p<0.0001) and of LDL-C in both genders (men p<0.05, women p<0.0001). The prevalence of overweight, obesity and central obesity were significantly higher in urban than in rural areas in both men and women. Compared to lean subjects, obese men and women had significantly higher mean serum TC, TG, LDL-C and a higher prevalence of dyslipidaemia. The mean levels of TC, TG and LDL cholesterol increased across successive increases in BMI and waist circumference quintiles in both genders. CONCLUSION: Subjects from the urban area had greater lipid abnormalities related to obesity than those from the rural area and that, central adiposity had a greater effect on total cholesterol and LDL cholesterol among women than was BMI.

Cardiovascular risk factors in Tanzania: a revisit.

In this assessment of cardiovascular risk factors, we examined the prevalence of selected risk factors according to the World Health Organisation (WHO) CARDIAC Study protocol and compared them with a similar study conducted more than a decade ago. The survey was carried out in Dar es Salaam (D, urban), Handeni (H, rural) and Monduli (Mo, semi-nomadic area). Subjects aged 47-57 were recruited randomly for blood pressure and anthropometrical measurements, 24 h urine collection and blood sampling. A structured questionnaire was used to obtain dietary information. The 1998 survey studied 446 subjects, while the 1987 survey included 496 men and women. The measured weight, body mass index (BMI) and prevalence of obesity (BMI > or = 30 kg/m(2)) increased significantly among women in the 1998 survey in rural Handeni and urban Dar. The overall prevalence of obesity was higher for women in the most recent survey (22.8%, P < 0.0001). Diastolic blood pressure (DBP) was higher in the most recent survey for women in Handeni. The overall prevalence of hypertension (blood pressure > 160/95 mmHg, or antihypertensive drug use), rose to 41.1% in 1998, (P < 0.001) for men and to 38.7% (P < 0.05) for women. The mean total serum cholesterol and prevalence of hypercholesterolaemia increased significantly in the most recent survey in the three studied areas. The overall prevalence of hypercholestrolaemia (serum cholesterol > 5.2 mmol/l) was higher in the 1998 survey for both men (21.8%, P < 0.0001) and women (54.0%, P < 0.0001). The mean HDL cholesterol increased significantly in the most recent survey, with a significant reduction in the mean atherogenic index, though these were still at higher levels (men 5.8, P < 0.0001; women 5.1, P < 0.0001 vs. 1987). A strong positive correlation was observed between blood pressure (SBP and DBP) and body mass index, total serum cholesterol and sodium to potassium ratio. These data suggest that for the past decade there has been an increase in the mean levels and prevalence of selected cardiovascular risk factors in Tanzania.

The relation of oxidative DNA damage to hypertension and other cardiovascular risk factors in Tanzania.

OBJECTIVES: To clarify the mechanism of involvement of oxidative stress in hypertensives, we investigated the relationship between the marker of oxidative DNA damage, urinary 8-hydroxy-2'-deoxyguanosine (8-OHdG), and cardiovascular risk factors, such as hypertension and serum glycosylated hemoglobin (HbA1c), among Tanzanians aged 46-58 years who were not on antihypertensive medication. DESIGN AND METHODS: Sixty subjects (males/females, 28/ 32) were selected randomly from the subjects who completed a 24h urine collection in our epidemiological study at Dar es Salaam, Tanzania in 1998. The subjects were divided into two groups, hypertensive subjects (systolic blood pressure (SBP) > or = 140 mmHg and/or diastolic blood pressure (DBP) > or =90 mmHg) and normotensive subjects (SBP < 140 mmHg and DBP < 90 mmHg) or hyperglycemic subjects (HbA1c > or = 6.0%) and normoglycemic subjects (HbA1c < 6.0%). Biological markers from urine and blood were analyzed centrally in the WHO Collaborating Center. RESULTS: The mean levels of HbA1c and 8-OHdG were significantly higher in the hypertensive subjects than in the normotensive subjects (P < 0.05). Urinary 8-OHdG was significantly higher in hyperglycemic subjects than in normoglycemic subjects. HbA1c was positively correlated with the 24-h urinary 8-OHdG excretions (r= 0.698, P < 0.0001). CONCLUSIONS: These findings suggest oxidative DNA damage is increased in hypertensive subjects, and there is a positive correlation between the level of blood glucose estimated as HbA1c and oxidative DNA damage. Hyperglycemia related to insulin resistance in hypertension in Tanzania is associated with increased urinary 8-OHdG.

Abnormalities in dihomo-gamma-linolenic acid release in the pathogenesis of hypertension.

Spontaneously hypertensive rats (SHR) respond to angiotensin and norepinephrine with an exaggerated pressor response. We have investigated the possibility that increased vascular reactivity in SHR may be related to a reduced synthesis of prostaglandin E1 (PGE1) resulting from a defect in the release of its precursor, dihomo-gamma-linoleic acid (DGLA). Isolated perfused mesenteric vascular beds of SHR and age matched Wistar-Kyoto rats (WKY) were perfused with Kreb's bicarbonate buffer. The effluent was collected and the fatty acid composition determined by gas chromatography. In SHR the release of DGLA, arachidonic acid, eicosapentaenoic acid, and virtually all other fatty acids detected in the effluent were reduced when compared to their normotensive controls. This difference could not be explained by low tissue fatty acid levels because these were higher in SHR. Evening primrose oil (EPO) when added to the diet increased the release of DGLA but not of other prostanoid precursors. EPO also reduced vascular reactivity and reduced blood pressure in SHR. It is suggested that the defect in the release of DGLA may be involved in the pathogenesis of hypertension because it occurs early before hypertension has actually occurred.

Ethnic differences in salt sensitivity: genetic or environmental factors?

1. The CARDIAC study, a world-wide cross-sectional epidemiological study on the relationship between alimentary factors and cardiovascular diseases, provided the initial evidence of ethnic differences in salt sensitivity; this was because despite much less urinary sodium excretion in Tanzania than in Brazil and the Japanese, the prevalence of hypertension was relatively higher in Tanzania than in the latter two populations. 2. To investigate this difference in salt sensitivity, a standardized clinical experiment was carried out: six to 13 male volunteers were placed on 2500 kcal basal diets containing 3 g salt/day. Eighteen grams of salt were added daily from the sixth to the 11th day in Tanzania and Brazil, while 22 g of salt were added in Japan. 3. Salt loading induced a significant rise in systolic blood pressure (SBP) on the second day of the high salt period (HSP) in Japan, the second and third day of HSP in Brazil, and all days of HSP in Tanzania. 4. Salt sensitivity was seen in 16.7% of the participants in Japan, 36.4% in Brazil and 46.2% in Tanzania. Further analysis of the effect of salt on blood pressure (BP) was carried out using the data from the CARDIAC study by multiple regression analysis. A within-centre comparison of fatty acid was also made. 5. The regression analysis revealed that the relationship of salt and blood pressure is more positively tight in Tanzania than Brazil and Japan after controlling for other confounding variables. Fatty acids in serum phospholipids contain significantly more palmitic acid and showed lower P/S ratios than those from Brazil and Japan.

The cardiac study in Tanzania: salt intake in the causation and treatment of hypertension.

The acute effects of varying salt intakes on the level of blood pressure in male normotensive volunteers in Tanzania were investigated. The subjects were kept on a diet supplying about 100 g protein, 85 g fat, and 2700 kcal per day. Daily potassium intake was about 50 mmol. The high salt diet increased urinary sodium excretion to 320 mmol/day while the low salt diet reduced sodium excretion to 52 mmol/day. Within 4-5 days there was a significant difference in mean arterial pressure between subjects on a high and a low salt diet. It is concluded that blood pressure in normotensive Tanzanian blacks is sensitive to alterations in salt intake. Although salt intake in Tanzania is relatively low, salt may be important in the causation of hypertension, and a reduction in salt intake may still have a place in the treatment and prevention of hypertension in Tanzania.

Diet and hypertension in Tanzania.

A survey was conducted on an urban population in the city of Dar es Salaam and on a rural community in both Handeni and the pastoral Masai in Monduli to investigate the relationship between diet and hypertension in Tanzania. Blood pressure (BP) was measured using an automatic BP-measuring machine. Biological markers of dietary intake were measured in 24-h urine and in blood. Hypertension was noted to be a bigger problem in the capital city, where the rate of obesity and salt intake were higher whereas potassium, protein, and polyunsaturated fatty acid intake were lower. Therefore, attention to dietary habits may reduce the growing problem of hypertension in Tanzania.

Release of fatty acids by perfused vascular tissue in normotensive and hypertensive rats.

The release of fatty acids from perfused mesenteries of spontaneously hypertensive rats (SHR) and control Wistar-Kyoto rats (WKY) was studied. The release of the prostaglandin precursors dihomogammalinolenic acid, arachidonic acid, and eicosapentaenoic acid was reduced in SHR when compared with age-matched WKY. The release of all other fatty acids detected in the effluent was also reduced. The differences in fatty acid release were evident even when tissue levels of the fatty acids were similar or higher in SHR than in controls. The addition of evening primrose oil and fish oil into the diet partially corrected these defects. Evening primrose oil and fish oil both attenuated increases in blood pressure, but fish oil was more potent than primrose oil. Although both diets reduced vascular reactivity, primrose oil was more effective with lower doses of norepinephrine whereas fish oil blunted the effects of both low and high doses of norepinephrine. The possible mechanisms for the effects of primrose oil and fish oil on vascular reactivity are briefly discussed.

The effects of indomethacin and PGE2 on vascular reactivity in spontaneously hypertensive rats: possible role of prostaglandins in the pathogenesis of hypertension.

Vascular reactivity was investigated in isolated perfused mesenteric vessels of young and adult SHR and age matched WKY. The primary objective was to investigate whether the difference in vascular reactivity between SHR and WKY would persist if endogenous prostaglandin synthesis was abolished and vascular reactivity restored with fixed exogenous quantities of PGE2. In young rats, when similar concentrations of PGE2 are infused in indomethacin blocked preparations, the difference in vascular reactivity between SHR and WKY is abolished. However, in adult rats the difference persists. It is concluded that enhanced prostaglandin synthesis in hypertension may participate in enhancing vascular reactivity.

Zinc and vascular reactivity in rat mesenteric vessels: possible altered dihomo-gamma-linolenic acid metabolism in spontaneously hypertensive rats.

Zinc at a concentration of 0.4 microgram/ml potentiated pressor responses to norepinephrine in isolated perfused mesenteric vessels of SHR and WKY. At a higher concentration, 3.2 micrograms/ml, it inhibited responses to norepinephrine in WKY but produced no such inhibition in SHR. However, a transient potentiation was observed in SHR with the higher concentration. Pressor responses to potassium in WKY were not affected by zinc at either concentration. In SHR, however, the higher dose of zinc inhibited pressor responses to potassium. The low dose had no effect. Since effects of zinc may be mediated by release of DGLA, we suggest that in SHR DGLA release may be impaired.

The roles of prostaglandins and calcium accumulation in muscular dystrophy.

There is good evidence that abnormal calcium accumulation may be a final common pathway of muscle degeneration in the muscular dystrophies. Prostaglandins are able to promote calcium entry into cells and excess prostaglandin activity coupled with a defect in intracellular calcium release could cause toxic accumulations of calcium in intracellular organelles such as mitochondria. Serotonin stimulates prostaglandin synthesis while tricyclic antidepressants inhibit calcium release from intracellular organelles thus possibly accounting for the models of muscular dystrophy reported using this combination. The prostaglandin/calcium hypothesis can account for the effects of vitamin E, steroids and local anaesthetic-like drugs in muscular dystrophy. Since many drugs already in clinical use for other purposes can be used to control prostaglandin synthesis or action this hypothesis has immediate potential clinical applications.

Actions of the tricyclic antidepressant clomipramine on responses to pressor agents. Interactions with prostaglandin E2.

Clomipramine inhibited pressor responses to potassium ions and vasopressin in the rat mesenteric vascular bed with an ID50 of about 1.8 microgram/ml against both pressor agents and the actions of indomethacin and PG2 on the clomipramine effect suggested that the drug may have been antagonising the action of an endogenous PG. This was supported by the inhibitory action of clomipramine on PG2 actions on guinea-pig ileum. A lower concentration also inhibited pressor responses to noradrenaline and angiotensin (ID50 about 9 ng/ml): inhibition was increased by PG2 and reduced by indomethacin. In this preparation potassium and vasopressin act primarily by stimulating calcium entry from the extracellular fluid whereas noradrenaline and angiotensin act primarily by releasing calcium from intracellular or membrane-bound stores. Our results can be explained by two actions: 1. a PG-antagonist action of clomipramine at the cell membrane and 2. a selectve inhibitory effect on release of intracellular calcium. Clomipramine may prove useful in studying PG and calcium-dependent mechanisms.

Prostaglandin A2 at low rates of infusion restores the antidiuretic effect of vasopressin in lithium-treated rats.

To test the effect of prostaglandin A2 (PGA2) on renal function, infusions of PGA2 (0-7 ng/kg/min), arginine-vasopressin (AVP) (1-25 ng/kg/min) and PGA2 plus AVP were administered to male rats made resistant to the antidiuretic effect of AVP by pre-treatment with lithium. In non-lithium-treated control rats, AVP had its expected antidiuretic action but in lithium-treated rats neither urinary volume nor osmolarity was changed. Prostaglandin A2 alone had no effect on urine output in lithium-treated rats; AVP plus PGA2 infused together evoked a near normal antidiuretic response. This antidiuretic action of PGA2 contrasts with the diuretic action reported by others. However, our infusion rates were 300-4000 times lower than those of other workers and it is suggested that PGs may have opposite actions on the kidney depending on their concentration. The effect of indomethacin (a blocker of prostaglandin synthesis) on urine flow was tested in five groups of rats on different régimes of liquid intake. Urine flow was reduced in the three groups with the highest urine volumes before treatment, and increased in the two groups with the lowest urinary volumes, again indicating that PGs may have both diuretic and antidiuretic actions.

Effects of prostaglandins on baseline pressure and responses to noradrenaline in a perfused rat mesenteric artery preparation. PGE1 as an antagonist of PGE2.

Concentrations of PGE1, PGE2, PGA1, PGA2 and PGF2alpha ranging from 10 to 10(5) pg/ml (2.8 x 10(-11) to 2.8 x 10(-7)M) were perfused through a rat mesenteric vascular bed preparation and their effects on baseline pressure and responses to noradrenaline noted. PGs A2 and FEalpha elevated baseline pressure at concentrations of 100 pg/ml and above but the other three PGs had little or no effect. PGs E2, A1 and F2alpha markedly potentiated responses to noradrenaline at all concentrations studied. In contrast PGs E1 and A2 caused a clear potentiation at 10 pg/ml but at higher concentrations the effect disappeared and concentrations of above 10(4) pg/ml were actually inhibitiory. E1 could antagonize the vascular effects of E2. These results offer possible explanations for some of the confusing findings reported in the literature.

Vascular actions of frusemide and bumetanide on the rat superior mesenteric vascular bed: interactions with prostaglandins.

1. The addition of frusemide or bumetanide to perfusion fluid inhibited the response of the isolated mesenteric vascular bed to noradrenaline. 2. Addition of prostaglandin E2 to the perfusion fluid completely restored the response to noradrenaline. 3. Inhibition of prostaglandin secretion by indomethacin with restoration of responses to noradrenaline by the addition of exogenous prostaglandin E2 prevented the inhibitory effect of frusemide or bumetanide on responses to noradrenaline. 4. The inhibitory effects of diuretics on responsiveness to noradrenaline is mediated by blockade of endogenous prostaglandin synthesis.

Cortisol in physiological concentrations acts within minutes to modify effects of prolactin and growth hormone on prostaglandin secretion: importance of effect in modulating cellular responses to calcium and cyclic nucleotides.

We propose that intracellular prostaglandins (PGs) are essential for the final expression of the effects of second messengers in most cells. We suggest that the amounts of PGs required are very small (in the picomolar range) and are much lower than those used in most current PG studies. We suggest that while therapeutic levels of inhibitors of PG synthetase may be adequate to block the overflow of PGs from cells, they are in most cases unlikely to reduce intracellular PGs sufficiently to test the role of such PGs. We propose that there is a basal level of PG synthesis unaffected by hormones but that above this level PG synthesis is regulated by the interplay between physiological levels of cortisol, prolactin, growth hormone and thyroid hormones. For the most part prolactin seems to stimulate PG synthesis and cortisol to inhibit it: cortisol has, however, no inhibitory effect on basal PG synthesis. In reducing prolactin-stimulated PG synthesis cortisol is 1000-2000 times more potent than indomethacin on a molar basis. We suggest that the regulation of intracellular PG levels is the mechanism of the so-called "permissive" actions of these hormones. These concepts could prove important in the understanding of many aspects of physiology and pathophysiology including diurnal and seasonal changes in hormone responsiveness. They are also relevant to the use of established drugs and the design of new ones.