Search

Background. This study aimed to identify correlates of case fatality within an incident stroke population in rural Tanzania. Methods. Stroke patients, identified by the Tanzanian Stroke Incidence Project, underwent a full examination and assessment around the time of incident stroke. Records were made of demographic data, blood pressure, pulse rate and rhythm, physical function (Barthel index), neurological status (communication, swallowing, vision, muscle activity, sensation), echocardiogram, chest X-ray and computed tomography (CT) head scan. Cases were followed up over the next 3 - 6 years. Results. In 130 incident cases included in this study, speech, language and swallowing problems, reduced muscle power, and reduced physical function were all significantly correlated with case fatality at 28 days and 3 years. Age was significantly correlated with case fatality at 3 years, but not at 28 days post-stroke. Smoking history was the only significant correlate of case fatality at 28 days that pre-dated the incident stroke. All other significant correlates were measures of neurological recovery from stroke. Conclusions. This is the first published study of the correlates of post-stroke case fatality in sub-Saharan Africa (SSA) from an incident stroke population. Case fatality was correlated with the various motor impairments resulting from the incident stroke. Improving poststroke care may help to reduce stroke case fatality in SSA.

OBJECTIVE--To compare the prevalence of electrocardiographic abnormalities and to evaluate the association between these abnormalities and the levels of coronary heart disease among Chinese living in different environments. DESIGN--Cross sectional surveys. SETTING--Beijing, China, and the island of Mauritius. SUBJECTS--Random samples of people aged 35-64 years in Beijing (621 men, 642 women) in 1984 and in Mauritius among Chinese (137 men, 130 women) and non-Chinese (1265 men, 1432 women) in 1987. MAIN OUTCOME MEASURES--Prevalence of electrocardiographic abnormalities suggesting coronary heart disease and of associated risk factors. RESULTS--Prevalence of electrocardiographic abnormalities suggesting coronary heart disease was significantly lower in Beijing (4.0%) than in Mauritian Chinese (24.3%) and Mauritian non-Chinese (24.5%). Mean serum concentrations of total and non-high density lipoprotein cholesterol were lower in Beijing Chinese than in Mauritian Chinese, but smoking and hypertension were slightly more prevalent. Overall, men with electrocardiographic abnormalities had higher risk factor levels than those with a normal electrocardiogram regardless of ethnic origin. CONCLUSIONS--The prevalence of coronary heart disease and associated risk factors was different among Chinese living in two different environments: in Beijing in the People's Republic of China and in Mauritius. Chinese, who traditionally have a very low frequency of coronary heart disease, are by no means protected against coronary heart disease and other non-communicable diseases. Therefore, primary prevention of coronary heart disease is a major challenge for preventive medicine in China, as well as in many other developing countries.

This article presents the conclusions of a WHO Expert Consultation that evaluated the utility of the ‘metabolic syndrome’ concept in relation to four key areas: pathophysiology, epidemiology, clinical work and public health. The metabolic syndrome is a concept that focuses attention on complex multifactorial health problems. While it may be considered useful as an educational concept, it has limited practical utility as a diagnostic or management tool. Further efforts to redefine it are inappropriate in the light of current knowledge and understanding, and there is limited utility in epidemiological studies in which different definitions of the metabolic syndrome are compared. Metabolic syndrome is a pre-morbid condition rather than a clinical diagnosis, and should thus exclude individuals with established diabetes or known cardiovascular disease (CVD). Future research should focus on: (1) further elucidation of common metabolic pathways underlying the development of diabetes and CVD, including those clustering within the metabolic syndrome; (2) early-life determinants of metabolic risk; (3) developing and evaluating context-specific strategies for identifying and reducing CVD and diabetes risk, based on available resources; and (4) developing and evaluating population-based prevention strategies.

This paper describes rates and causes of injury deaths among community members in three districts of the United Republic of Tanzania. A population-based study was carried out in two rural districts and one urban area in Tanzania. Deaths occurring in the study areas were monitored prospectively during a period of six years. Censuses were conducted annually in the rural areas and biannually in the urban area to determine the denominator populations. Cause-specific death rates and Years of Life Lost (YLL) due to injury were calculated for the three study areas. During a 6 year period (1992-1998), 5047 deaths were recorded in Dar es Salaam, 9339 in Hai District and 11 155 in Morogoro Rural District. Among all ages, deaths due to injuries accounted for 5% of all deaths in Dar es Salaam, 8% in Hai and 5% in Morogoro. The age-standardised injury death rates among men were approximately three times higher than among women in all study areas. Transport accidents were the commonest cause of mortality in all injury-related deaths in the three project areas, except for females in Hai District, where it ranked second after intentional self-harm. We conclude that injury deaths impose a considerable burden in Tanzania. Strategies should be strengthened in the prevention and control of avoidable premature deaths due to injuries.

Little research has been conducted on health in Chinese communities in the UK and there are few representative data on smoking, alcohol consumption or other aspects of lifestyle. We undertook a cross sectional population-based study of 380 Chinese and 625 European men and women aged 25 to 64 y, using self-completion and interview questionnaires in Newcastle upon Tyne, UK between 1991 and 1995. We measured self-reported prevalence of cigarette smoking, number of cigarettes smoked per week and age at starting smoking; self-reported prevalence of alcohol consumption and units of alcohol consumed per week. In age-adjusted comparisons smoking was less common in Chinese (24%) than European men (35%) (P=0.00002) and among Chinese (1%) compared with European women (33%) (P0.00001). Number of cigarettes smoked was similar among Chinese and European male smokers. Median age at starting smoking was higher among Chinese (18.5 y) compared with European men (15 y) (P=0.00001). Smoking was commonest in older Chinese and in younger Europeans. The prevalence of alcohol consumption was lower among Chinese (63%) than European men (93%) (P0.00001) and among Chinese (29%) compared to European women (89%) (P0.00001). Median alcohol consumption was significantly lower among Chinese (2 units/week) than European men (16 units/week) (P0.00001), and among Chinese (1 unit/week) compared to European women (6 units/week) (P0.00001). Among those who drank alcohol, Chinese men were less likely to drink above recommended limits than European men (1% vs 39%; P0.00001). Chinese men and women currently have relatively favourable patterns of smoking compared to European adults in Newcastle. Average alcohol consumption among Chinese who drink is lower than among Europeans, and a substantial proportion of the Chinese population in Newcastle drink no alcohol. Patterns of health related behaviour should be tracked over time in ethnic minority populations to identify changes that pose risk to health and which deserve appropriate intervention.

There is evidence that the prevalence of certain non-communicable diseases, such as diabetes and hypertension, is increasing rapidly in parts of sub-Saharan Africa. Others, such as asthma and epilepsy, are known to be common but to be poorly managed. This paper describes a project, funded by the Department for International Development of the British Government, which aims to provide costed and evaluated treatment packages for use at primary health care level, methods and materials for evaluating the quality of non-communicable disease care, and a protocol for the assessment for national opportunities for the prevention of hypertension, heart disease and diabetes. Methods are being developed and piloted in urban and rural Tanzania and Cameroon.Declining death rates from communicable diseases, together with population aging, leads to a higher incidence and prevalence of noncommunicable diseases (NCDs), such as atherosclerotic disorders, cancers, and chronic respiratory disease. These NCDs gradually become the population's predominant health problems. Evidence indicates that the prevalence of certain NCDs, such as diabetes and hypertension, is increasing rapidly in parts of sub-Saharan Africa. Others, such as asthma and epilepsy, are common, but poorly managed. This paper describes a project funded by the British Government's Department for International Development to provide costed and evaluated treatment packages for use at the primary health care level, methods and materials for evaluating the quality of noncommunicable disease care, and a protocol for assessing national opportunities to prevent hypertension, heart disease, and diabetes. Methods are now being developed and piloted in urban and rural Tanzania and Cameroon.

Most data for stroke mortality in sub-Saharan Africa are hospital based. We aimed to establish the contribution of cerebrovascular disease to all-cause mortality and cerebrovascular disease mortality rates in adults aged 15 years or more in one urban and two rural areas of Tanzania.Regular censuses of the three surveillance populations consisting of 307,820 people (125,932 aged below 15 years and 181,888 aged 15 or more) were undertaken with prospective monitoring of all deaths arising in these populations between June 1, 1992 and May 31, 1995. Verbal autopsies were completed with relatives or carers of the deceased to assess, when possible, the cause of death.During the 3-year observation period 11,975 deaths were recorded in the three surveillance areas, of which 7629 (64%) were in adults aged 15 years or more (4088 [54%] of these in men and 3541 [46%] in women). In the adults, 421 (5.5%) of the deaths were attributed to cerebrovascular disease, 225 (53%) of these in men and 196 (47%) in women. The yearly age-adjusted rates per 100,000 in the 15-64 year age group for the three project areas (urban, fairly prosperous rural, and poor rural, respectively) were 65 (95% CI 39-90), 44 (31-56), and 35 (22-48) for men, and 88 (48-128), 33 (22-43), and 27 (16-38) for women, as compared with the England and Wales (1993) rates of 10.8 (10.0-11.6) for men and 8.6 (7.9-9.3) for women.We postulate that the high rates in Tanzania were due to untreated hypertension. Our study assessed mortality over a single time period and therefore it is not possible to comment on trends with time. However, ageing of the population is likely to lead to a very large increase in mortality from stroke in the future.

Three schemes are presented for discussion whereby physicians undergoing postdoctoral training can combine a period of research training with their clinical training and so enable those who wish to follow a career in academic medicine to do so, or alternatively to revert to a clinical career. The training arrangements for those wishing to take up clinical academic medicine have hitherto been uncertain and hence unattractive to some. As well as encouraging more high-calibre trainees into academic medicine, the training programmes described are intended to bring greater clarity to those responsible for academic and clinical training and to those who fund research.

To investigate whether relative baseline leptin levels predict long-term changes in adiposity and/or its distribution.In a longitudinal study of 2888 nondiabetic Mauritians aged 25 years to 74 years who participated in population-based surveys in 1987 and 1992, changes in body mass index (BMI), waist/hip ratio (WHR), and waist circumference were compared between "hyperleptinemic," "normoleptinemic," and "hypoleptinemic" groups. "Relative leptin levels" were calculated as standardized residuals from the regression of log10 leptin on baseline BMI to provide a leptin measure independent of BMI. Analyses were performed within each sex. A linear regression model was used to assess the effect of standardized residuals on changes in BMI, WHR, and waist circumference, independent of baseline BMI, age, fasting insulin, and ethnicity.After adjusting for age and baseline BMI by analysis of covariance, there was no difference in changes in BMI, WHR, or waist circumference between men with low, normal, or high relative leptin levels. Among women, there was a significant difference in deltaWHR across leptin groups, such that the largest increase occurred in the "normal" leptin group. For both men and women, the linear regression models explained approximately 10% of variation in dependent variables, and the only significant independent variables were age, BMI, and being of Chinese origin, compared with Indian origin.These findings do not support a role for leptin concentration in predicting weight gain or changes in fat distribution in adults over a 5-year period.

The classification of diabetes mellitus and the tests used for its diagnosis were brought into order by the National Diabetes Data Group of the USA and the second World Health Organization Expert Committee on Diabetes Mellitus in 1979 and 1980. Apart from minor modifications by WHO in 1985, little has been changed since that time. There is however considerable new knowledge regarding the aetiology of different forms of diabetes as well as more information on the predictive value of different blood glucose values for the complications of diabetes. A WHO Consultation has therefore taken place in parallel with a report by an American Diabetes Association Expert Committee to re-examine diagnostic criteria and classification. The present document includes the conclusions of the former and is intended for wide distribution and discussion before final proposals are submitted to WHO for approval. The main changes proposed are as follows. The diagnostic fasting plasma (blood) glucose value has been lowered toor =7.0 mmol l(-1) (6.1 mmol l(-1)). Impaired Glucose Tolerance (IGT) is changed to allow for the new fasting level. A new category of Impaired Fasting Glycaemia (IFG) is proposed to encompass values which are above normal but below the diagnostic cut-off for diabetes (plasmaor =6.1 to7.0 mmol l(-1); whole bloodor =5.6 to6.1 mmol l(-1)). Gestational Diabetes Mellitus (GDM) now includes gestational impaired glucose tolerance as well as the previous GDM. The classification defines both process and stage of the disease. The processes include Type 1, autoimmune and non-autoimmune, with beta-cell destruction; Type 2 with varying degrees of insulin resistance and insulin hyposecretion; Gestational Diabetes Mellitus; and Other Types where the cause is known (e.g. MODY, endocrinopathies). It is anticipated that this group will expand as causes of Type 2 become known. Stages range from normoglycaemia to insulin required for survival. It is hoped that the new classification will allow better classification of individuals and lead to fewer therapeutic misjudgements.

The American Diabetes Association have recommended that the fasting plasma glucose level for the diagnosis of diabetes is lowered and that this becomes the main diagnostic test. We have used population-based data from three ethnic groups in Newcastle upon Tyne to examine the implications of this change. Data were available on 824 European (25-74 years), 375 Chinese (25-64 years), and 680 South Asian (25-74 years) subjects. All subjects apart from those reporting a prior diagnosis of diabetes underwent a standard 75 g oral glucose tolerance test (WHO criteria) which included the measurement of fasting glucose. The prevalence of diabetes was higher in all three ethnic groups using the new ADA criteria compared to the WHO criteria: 7.1% vs 4.8% in Europeans; 6.2% vs 4.7% in Chinese; and 21.4% vs 20.1% in South Asians. There was much variation in individuals categorized by the ADA and WHO criteria. Agreement between the two for the diagnosis of previously unknown diabetes was only moderate (kappa statistics 0.42 to 0.59). Thus in the populations studied the new criteria would increase the prevalence of diabetes in addition to classifying some individuals diabetic by current criteria as non-diabetic. It should be stressed however that diagnosis of the individual should not be based on a single test.

A cross-sectional study was conducted on a 20-74-year-old population in an urban white-Hispanic population in Paraguay to determine the prevalence of diabetes mellitus (DM), impaired glucose tolerance (IGT), and associated cardiovascular disease (CVD) risk factors. In total 1606 subjects completed the study (response rate 80.3%; 1094 women, 512 men). The overall prevalences were: DM 6.5%, IGT 11.3%, hypertension 17.1%, and obesity 31.6% with more obesity in women (35.7% vs 22.8%, p0.05). Age-standardized prevalences were: DM 6.5%, IGT 13.5% in females and DM 5.5%, IGT 7.2% in males. DM and IGT subjects had two or more CV risk factors significantly more often than the normal population. In conclusion, DM, IGT, hypertension, and obesity are common in this South American Hispanic urban population, particularly in women. Public health measures, such as lifestyle education, are required to decrease these noncommunicable diseases.

The relative contribution of socioeconomic, behavioural and biological factors operating in fetal and infant life, childhood and adulthood to risk for cardiovascular disease, respiratory diseases and non-insulin-dependent diabetes in middle age has become an important research issue. All 1142 babies born in Newcastle upon Tyne in May and June 1947 were recruited into a prospective cohort study of child health (the 'Thousand Families' study) and followed in great detail to the age of 15 y, with a brief further follow up at age 22 y. Children from poorer families were at greatest risk of severe respiratory tract infection in infancy. Children from professional and managerial families were on average taller and heavier throughout childhood than those from semi- and unskilled manual social classes. Repeated infections in early childhood greatly increased the risk of developing chronic respiratory disease by age 15 y. This paper outlines a new investigation designed to trace surviving members of this cohort and to chart the relationships between their socioeconomic circumstances, lifestyles, experiences and health from birth through to the present day. Existing data on socioeconomic circumstances and infections in infancy and childhood, infant nutrition, birthweight and physical development to age 22 y will be linked to information gained from a new study. This comprises a postal questionnaire survey of study members' adult health, socioeconomic circumstances and lifestyle, and a hospital based clinical examination including heart and lung function, glucose tolerance, blood lipids and anthropometric measurements at age 49-51 y. Out of a target sample of 979 people for whom sufficient data are available on the first year of life, 866 (88%) have been traced and 649 are still resident in the North of England. Those study members who have been traced are highly representative of the original cohort. The Thousand Families cohort provides a unique opportunity for detailed epidemiological study because of the wealth of data available on infant and childhood socioeconomic and family circumstances, all of which was collected prospectively. In addition, there has been comparatively little loss to follow-up since 1948.

It has been shown previously in smaller studies that fasting serum leptin and insulin concentrations are highly correlated, and insulin sensitive men have lower leptin levels than insulin resistant men matched for fat mass. We have examined the association between insulin resistance (assessed by fasting insulin) and leptin after controlling for overall and central adiposity in a population-based cohort.Leptin levels were compared across insulin resistance quartiles within three categories of obesity (tertiles of body mass index (BMI)). Partial correlation coefficents and multiple linear regression models were used to assess the relationship between leptin and fasting insulin after adjusting for BMI and waist to hip ratio (WHR) or waist circumference.Subjects were normoglycemic participants of a 1987 non-communicable diseases survey conducted in the multiethnic population of Mauritius. 1227 men and 1310 women of Asian Indian, Creole and Chinese ethnicity had normal glucose tolerance and fasting serum leptin measurements.Mean serum leptin concentration increased across quartiles of fasting insulin in each BMI group and gender, after controlling for BMI, WHR and age. Furthermore, fasting insulin was a significant determinant of serum leptin concentration, independent of BMI and WHR, in both men and women. Similar results were found if waist circumference replaced BMI and WHR in the model.These results suggest that insulin resistance/concentration may contribute to the relatively wide variation in leptin levels seen at similar levels of body mass or alternatively, leptin may play a role in the etiology of insulin resistance. Further studies will be important to determine whether the hyperleptinemia/insulin resistance relationship has a role in the natural history of obesity, Type 2 diabetes mellitus and the other metabolic abnormalities associated with insulin resistance.

There is a paucity of research on health in the UK Chinese community partly due to the difficulties of identifying and accessing study populations. For a survey of cardiovascular disease we aimed to identify and recruit all Chinese adults aged 25-64 y living in Newcastle-upon-Tyne, UK. One thousand, eight hundred and sixty-five potential subjects were identified using a variety of methods. Of the 1702 potential subjects identified from a name analysis of the 1991 FHSA register (FHSA group), 638 students in halls of residence were excluded and the remaining 1064 were invited to participate. Non-respondents were followed up. Of the 1064, 658 (65.5%) addresses were no longer valid, 21 (2%) were reclassified as non-Chinese and no contact was made with 18 individuals (1.6%). A further 163 subjects (non-FHSA group) came forward in response to publicity, giving a total of 530 Chinese actually identified in Newcastle. Three hundred and eighty subjects took part in the study. Compared to the 1991 Census, the recruitment procedure underestimated the total population size, particularly for men and younger ages. In the FHSA group, men were significantly more likely to be current drinkers, and women were more likely to smoke and have a lower educational attainment that the non-FHSA group. There were no other important differences in the distribution of CHD risk markers in the two groups. Our experience indicates that the FHSA register is suitable for identifying Chinese but should be used alongside other complementary methods to augment samples for ethnicity and health research.

Leptin's association with fasting insulin raises the possibility that hyperleptinaemia is an additional component of the Metabolic Syndrome, or perhaps underlies the syndrome. This population-based study of Western Samoans examined the relationship of serum leptin with insulin sensitivity assessed by Homeostatic Model Assessment (HOMA) and components of the Metabolic Syndrome. Two hundred and forty subjects (114 men, 126 women), aged 28-74 years, were drawn from a study conducted in 1991. An oral glucose tolerance test indicated that 59 subjects had diabetes. Diabetic men had higher leptin levels than non-diabetic (6.0 vs 3.2 ng ml-1) but this difference was no longer significant after adjustment for BMI. Leptin levels in diabetic women (24.7 ng ml-1) non-diabetic women (22.6 ng ml-1) were not different. Leptin was strongly, positively correlated with BMI, fasting insulin and mean blood pressure after adjusting for age and sex (r0.43, p0.001), irrespective of glucose tolerance status. Linear regression models indicated that leptin was associated with insulin sensitivity independent of age, BMI, waist/hip ratio, triglycerides, HDL-cholesterol, and hypertension. Similar models were computed with mean blood pressure or triglycerides as the dependent variable, and including insulin sensitivity with the independent variables. Leptin was independently associated with mean blood pressure in men, but was not independently associated with triglycerides. Mean levels of 2-h insulin, triglycerides, LDL-cholesterol, and systolic blood pressure varied across tertiles of leptin in men after adjusting for age, BMI, and insulin sensitivity, and mean levels in the top tertile tended to be higher than in the lowest tertile. These results indicate an independent relationship between leptin and insulin sensitivity, but the equivocal results concerning associations of leptin with components of the Metabolic Syndrome make it unlikely that leptin affects these directly.

Impaired glucose tolerance (IGT) was introduced in 1979 as an intermediate category covering the grey area between unequivocal diabetes mellitus and risk free more normal glucose tolerance. The IGT group included those at high risk of subsequent development of non-insulin-dependent diabetes mellitus (NIDDM) but low risk of specific diabetic complications. Categorisation of subjects as IGT is hampered by the variability of the oral glucose tolerance test, but even those shown to be IGT only once are at increased risk of developing NIDDM. The relative roles of inheritance, fetal undernutrition, and environmental life style factors (physical inactivity and diet) in the aetiology and pathogenesis of IGT are discussed, with all contributing. The prevalence of IGT in different populations has now been widely studied with values ranging from 2 to 25% in adults. Rates of progression to NIDDM also vary widely from 2 to 14% per year. Risk factors for progression are discussed. IGT also carries an increased risk of development of cardiovascular disease (CVD) and forms part of the "metabolic syndrome". The role of insulin resistance as a common aetiological factor is briefly reviewed. Finally, possible means of treatment of IGT are listed with the intent of delaying the onset of diabetes and CVD, which is of obvious clinical importance.

Triglycerides are an important risk factor for coronary heart disease in Type 2 (non-insulin-dependent) diabetes mellitus. Although Type 2 diabetic patients have an exaggerated postprandial triglyceride response to a fat meal test, little is known about the variability of triglyceride concentrations in day-to-day life. We have studied the variability in triglyceride concentrations in 24 Type 2 diabetic patients over 6 months by having them record fasting and postprandial triglyceride concentrations at home using a Reflotron dry chemistry analyser. All patients were able to use the analyser effectively, with a correlation of 0.97 between patients' monthly Reflotron readings and those recorded by the laboratory. Over 1600 measurements were performed. The results demonstrate a large variation in both fasting (median 1.95 mmol l-1, range 0.8-6.7 mmol l-1) and postprandial triglyceride concentrations (median 2.68 mmol l-1, range 0.8-6.7 mmol l-1). This variation was accounted for by both a large intra- and inter-individual variation. Although there was a strong correlation overall between fasting and postprandial triglyceride concentrations r = 0.925 (p0.001), this did not apply on an individual basis. In conclusion, the large variability in triglyceride concentrations should be considered before introducing pharmacological therapy for hypertriglyceridaemia in Type 2 diabetes mellitus.

The aim was to establish whether risk factors for cardiovascular disease (CVD) are positively and independently associated with fasting insulin and/or body mass and waist-hip ratio in healthy elderly Nigerian subjects. Fasting plasma glucose, insulin, total cholesterol, triglycerides, blood pressure, and basal insulin resistance (HOMA method) were measured in 500 healthy elderly (or = 55 years) Nigerian volunteers (295 men, 205 women). Associations between blood pressure, triglycerides or cholesterol and fasting insulin, HOMA, body mass index (BMI) or waist-hip ratio were examined using linear regression. Age was controlled for in all analyses. In men, diastolic and systolic blood pressure were strongly associated with BMI, while there was no evidence of an independent relationship with fasting insulin or HOMA. Triglycerides were strongly associated with waist-hip ratio, with a weaker independent association with HOMA but not fasting insulin; fasting insulin and HOMA showed strong independent associations with total cholesterol. In women diastolic and systolic blood pressure were also strongly associated with BMI, but there was an independent relationship with fasting insulin for diastolic blood pressure and a less significant (p = 0.057) one for systolic blood pressure. Triglycerides were significantly associated with BMI but none of the other variables; there were no significant associations with cholesterol. There was no evidence of interaction between fasting insulin or HOMA and BMI or waist-hip ratio. The results suggest the hypotheses that in this population BMI or waist-hip ratio are stronger determinants of blood pressure and triglyceride levels than fasting insulin or HOMA, and that where insulin does play a role its effects are separate and additive.

There is evidence from animal models that postprandial insulin hypersecretion may precede the development of obesity and insulin resistance, but it is not clear if this is the case in humans. Recently, two longitudinal studies have suggested that insulin resistance acts to limit further weight gain rather than to promote it. The relationship of markers of insulin sensitivity and secretion to changes in weight and the waist to hip ratio (WHR) was therefore examined in nondiabetic Asian Indian (n = 2,169), Creole (n = 798), and Chinese (n = 189) Mauritians over a 5-year follow-up period. Younger age and lower initial body mass index (BMI) were consistent independent predictors of increase in weight in all sex-ethnic subgroups, and older age, higher BMI, and lower WHR were associated with change in WHR. Insulin sensitivity was assessed by homeostatic model assessment (HOMAS), as well as by fasting insulin and the ratio of fasting insulin to glucose. Insulin resistance predicted weight gain in Chinese men independently of baseline age and BMI. In Asian Indian and Creole men and women, these correlations were in the opposite direction (ie, insulin sensitivity predicted weight gain) but became nonsignificant when age and BMI were controlled. There was little relationship of insulin resistance/sensitivity to the change in WHR once baseline BMI was controlled. These data provide suggestive but not convincing evidence that insulin resistance may limit weight gain, and contradictory evidence in one ethnic group that insulin resistance promotes weight gain.

The aetiology of NIDDM is uncertain, although family and twin studies indicate an important role for genetic factors in disease onset. The function and position of IRS-1 within the insulin signalling pathway make it a prime candidate gene for the development of insulin resistance and NIDDM. Insulin resistant families were identified by studying unaffected first degree relatives from families with 2 or more living NIDDM subjects. Insulin sensitivity was determined in the relatives using the insulin tolerance test, and 15 families were identified as insulin resistant. One NIDDM subject from the 10 most resistant families was selected and the entire coding region of IRS-1 analysed by SSCP analysis. Four normoglycaemic subjects with no family history of diabetes served as controls. Five variant sequences of IRS-1 were identified with the NIDDM subjects; 2 silent polymorphisms at codons 235 (GGG to GGA) and 893 (CCG to CCC): 2 non-conservative mutations (Ala513Pro; Gly972Arg) and a codon deletion (Ser681-7 to Ser681-6). The influence of the non-conservative mutations alone, and in combination with other abnormalities of the insulin signalling pathway on peripheral insulin action, remains to be determined.

OBJECTIVE--To measure age and sex specific mortality in adults (15-59 years) in one urban and two rural areas of Tanzania. DESIGN--Reporting of all deaths occurring between 1 June 1992 and 31 May 1995. SETTING--Eight branches in Dar es Salaam (Tanzania's largest city), 59 villages in Morogoro rural district (a poor rural area), and 47 villages in Hai district (a more prosperous rural area). SUBJECTS--40,304 adults in Dar es Salaam, 69,964 in Hai, 50,465 in Morogoro rural. MAIN OUTCOME MEASURES--Mortality and probability of death between 15 and 59 years of age (45Q15). RESULTS--During the three year observation period a total of 4929 deaths were recorded in adults aged 15-59 years in all areas. Crude mortalities ranged from 6.1/1000/year for women in Hai to 15.9/1000/year for men in Morogoro rural. Age specific mortalities were up to 43 times higher than rates in England and Wales. Rates were higher in men at all ages in the two rural areas except in the age group 25 to 29 years in Hai and 20 to 34 years in Morogoro rural. In Dar es Salaam rates in men were higher only in the 40 to 59 year age group. The probability of death before age 60 of a 15 year old man (45Q15) was 47% in Dar es Salaam, 37% in Hai, and 58% in Morogoro; for women these figures were 45%, 26%, and 48%, respectively. (The average 45Q15s for men and women in established market economies are 15% and 7%, respectively.) CONCLUSION--Survivors of childhood in Tanzania continue to show high rates of mortality throughout adult life. As the health of adults is essential for the wellbeing of young and old there is an urgent need to develop policies that deal with the causes of adult mortality.

The WHO and the National Diabetes Data Group have produced definitions to classify subjects with blood glucose levels following oral ingestion that are higher than normal, but are less than for patients with diabetes. The term impaired glucose tolerance (IGT) has been used to describe this group who have an increased risk of developing non-insulin dependent diabetes mellitus and cardiovascular disease, compared with subjects with normal glucose tolerance. Wide intra-subject variability has been observed when glucose tolerance tests are carried out on separate occasions. This has led some researchers to suggest that IGT merely constitutes a blurred boundary between normality and diabetes but accumulating evidence now supports the hypothesis that IGT is a useful risk factor category.

Vibration perception threshold (VPT) was measured in 1185 Indian, Creole, and Chinese men and women in Mauritius, where the current prevalence of diabetes mellitus in adults aged 25-74 years is estimated to be 13%. Vibration perception threshold was measured using a biothesiometer, at seven sites in the hand, wrist, foot, and ankle, during a population survey in 76% of 574 known diabetic patients (KDM), 79% of 525 newly diagnosed diabetic patients (NDM), 18% of 1121 subjects with impaired glucose tolerance (IGT), and in 127 normal subjects. The association of VPT with glucose tolerance and other risk factors was assessed in order to identify individuals most at risk of foot ulceration and to determine whether risk factors and normal levels for VPT in these ethnic groups were consistent with those reported for Caucasians. After adjusting for age and height, geometric mean VPT at six of seven sites increased significantly with worsening glucose tolerance and increasing duration of diabetes in both men and women, VPT also increased significantly with level of fasting plasma glucose in men, but not women. Smoking and alcohol consumption had no effect on VPT, and body-mass index (BMI) was positively associated only at some sites. Chinese subjects had lower VPTs than Indians or Creoles. In multiple linear regression models, age, male gender, duration of diabetes, ethnic group, and height (lower extremity sites) were significantly associated with VPT among diabetic subjects.(ABSTRACT TRUNCATED AT 250 WORDS)

We investigated awareness of hypoglycaemia and its counter-regulatory hormone response in six young (ages 24-49, mean 30 years) and seven elderly (ages 65-80, mean 72 years) healthy non-diabetic subjects. A hyperinsulinaemic glucose clamp was used to control blood glucose level on two separate occasions. During the hypoglycaemic study, blood glucose was lowered in a stepwise manner to plateaus of 3.5, 3.0 and 2.5 mmol/l. A symptom score, visual reaction time test and digit symbol substitution test was completed pre-study, and at the end of each plateau. Pulse rate, blood pressure and counter-regulatory hormone measurements were taken every 15-30 min. The euglycaemic study was identical except that blood glucose remained at the fasting level. During hypoglycaemia, the elderly group had lower symptom scores than the young group (total relative score at glucose 2.5 mmol/l, mean +/- SEM: elderly -1 +/- 2.5, young 23 +/- 6.7, p0.01) and fewer individual symptoms despite a similar counter-regulatory hormone response. There was no difference in deterioration of the visual reaction times or digit symbol substitution scores during hypoglycaemia between the age groups. Unlike the young group, the elderly subjects had no tachycardia in response to hypoglycaemia. Their reduced awareness of hypoglycaemia may be due to an impaired end-organ response to counter-regulatory hormones, resulting in fewer symptoms.

Gliclazide is a second-generation sulfonylures that is widely used in the treatment of non-insulin-dependent diabetes mellitus (Type 2 diabetes). It has been recommended for use on the basis of both its metabolic and nonmetabolic effects. It has a clear beneficial effect on metabolic control in Type 2 diabetes. Blood glucose and lipid levels are lowered. The glucose-lowering effects are secondary to both enhanced insulin secretion and a decrease in insulin resistance. The former is due to closure of a K+ adenosine triphosphate (ATP) channel in the beta cell. The mechanism whereby insulin action on the liver and muscle are potentiated remains unknown. It does not appear to involve the insulin receptor, and although glycogen synthase activation is enhanced, this is probably not specified. It has proven difficult to separate the metabolic effects of gliclazide form the effects of improved control. The metabolic actions are probably also shared with over sulfonylureas. Gliclazide also has beneficial effects on platelet behavior and function and on the endothelium, in addition to improving free radical status. These effects should be beneficial for the prevention of diabetic microangiopathy. Some evidence has appeared for the prevention of deterioration of diabetic retinopathy, but results are variable and more convincing studies are required. Many of the nonmetabolic effects of gliclazide appear to be unique to this agent. Gliclazide thus appears to be a reasonable choice in the treatment of Type 2 diabetes with diet failure, both from the metabolic and non-metabolic standpoint.

Many studies of Type 2 (non-insulin-dependent) diabetes mellitus assume that the condition is homogeneous and clearly defined. There are, however, several problems with these assumptions. Thus, definition of Type 2 diabetes is one of exclusion of other types (insulin-dependent, malnutrition-related, gestational and other rarer types) and inevitably contains a heterogeneous group of disorders the aetiology of which is largely unclear, and separation from the insulin-dependent type can be problematic. Diagnosis is also imprecise in asymptomatic subjects due to the lack of accurate diagnostic tools and lack of clear distinction of impaired glucose tolerance. An alternative to the oral glucose tolerance test is urgently needed. Epidemiological studies of Type 2 diabetes and its complications are also fraught with difficulties due to variability of the oral glucose tolerance test, potential problems in glucose measurement, heterogeneity, population selection and problems in international comparisons due to differing age ranges and life expectancy. Great care is needed in all studies of Type 2 diabetes to ensure that the groups under study are properly selected, well-defined and fully described.

The prevalence of dyslipidaemia in children with insulin dependent diabetes mellitus (IDDM) and its relation to glycaemic control was studied in a group of 51 diabetic children and a control population of 132 schoolchildren. The prevalence of dyslipidaemia in the fasting state was increased in the diabetic group (39%) compared with control subjects (17%). Serum cholesterol concentration alone was raised in 25% of diabetic subjects while serum cholesterol and triglycerides were raised in 14%, compared with 16% and 0.7% respectively in control subjects. Serum total cholesterol (5.1 v 4.5 mmol/l), low density lipoprotein cholesterol (3.2 v 2.6 mmol/l), non-esterified fatty acids (0.91 v 0.50 mmol/l), and triglycerides (0.94 v 0.76 mmol/l) were higher in diabetic children. Serum total cholesterol, triglycerides, and apolipoprotein (apo)B concentrations increased with worsening control, while serum high density lipoprotein cholesterol and apoA-I concentrations were unaltered. There were also positive correlations between glycated haemoglobin and total cholesterol, triglycerides, and apoB in diabetic children. Thus, abnormalities in circulating lipids are common in young subjects with IDDM but largely disappear if blood glucose concentrations are reasonably controlled.

The aim of the study was to investigate whether the constellation of cardiovascular disease risk factors, described as Insulin Resistance Syndrome, exists in the multi-ethnic population of Mauritius, and to assess whether hyperinsulinaemia is the key feature of this syndrome. A sample of 5080 Mauritian subjects (aged 25-74 years) was examined in a noncommunicable diseases survey in 1987. Survey procedure included an oral glucose tolerance test, and anthropometric, blood pressure, plasma lipids and serum insulin measurements. Abnormal glucose tolerance (diabetes and impaired glucose tolerance), general obesity, upper-body obesity, hypertension, low HDL-cholesterol, and hypertriglyceridaemia were defined as risk factor conditions. Mean values for a series of risk factor variables were compared between reference subjects (no risk factors) and those with a risk factor condition (either one condition only, or in combination with one or more others). Prevalence estimates for each risk factor condition in combination with three or more other conditions were three to four times greater than expected by chance, and levels of risk factors for subjects with more than one risk factor condition were further away from the reference levels than for those with just one condition. Fasting and 2-h serum insulin levels were elevated for each condition when in combination with others, or to a lesser extent when isolated. However, this was not the case for isolated hypertension where insulin levels were not elevated. When adjusted for age, sex, and body mass index, insulin levels were only significantly elevated in subjects with upper-body obesity if in association with general obesity.(ABSTRACT TRUNCATED AT 250 WORDS)

Cirrhosis is characterized by paradoxical growth hormone secretion in response to glucose and insulin infusion. To ascertain whether this abnormality contributes to insulin resistance, euglycemic hyperinsulinemic glucose clamps were performed on six patients with cirrhosis and six normal control subjects. Each patient with cirrhosis underwent two clamps in random order, a clamp with somatostatin (250 micrograms/hr) together with insulin and glucagon replacement, and a control clamp without somatostatin. The normal subjects underwent the control clamp only. During the control clamp, growth hormone levels were considerably higher in the patients with cirrhosis (6.1 +/- 0.4 vs. 0.5 +/- 0.4 mU/L, p0.02), and glucose uptake was considerably lower (3.29 +/- 0.56 vs. 9.52 +/- 1.14 mg/kg/min, p0.001). Indirect calorimetry indicated that the defect was accounted for by lower nonoxidative glucose disposal (1.23 +/- 0.45 vs. 6.00 +/- 0.73, p0.001). Peripheral glucose uptake, exemplified by forearm glucose uptake (0.27 +/- 0.04 vs. 1.22 +/- 0.42 mg/100 ml/min, p0.02), and calculated insulin sensitivity (24 +/- 8 vs. 114 +/- 20 microliters/kg/min per mU/L) were particularly diminished. In the patients with cirrhosis somatostatin suppressed growth hormone levels (6.1 +/- 1.2 to 1.2 +/- 0.4 mU/L, p0.05). However, no significant changes occurred in whole-body glucose uptake (3.29 +/- 0.56 vs 3.01 +/- 0.54 mg/kg/min), forearm glucose uptake (0.27 +/- 0.04 vs 0.30 +/- 0.01 mg/100 ml/min) or insulin sensitivity (24 +/- 8 vs, 35 +/- 10 microliters/kg/min/mU/L, p = 0.42).(ABSTRACT TRUNCATED AT 250 WORDS)

Cardiovascular disease is two to three times more common in diabetic patients than in the non-diabetic population. Although risk factors that affect the general population such as age, cigarette smoking, hypertension, obesity and hypercholesterolaemia also affect diabetic subjects, the increased prevalence of hypertension and obesity in non-insulin-dependent diabetes mellitus (NIDDM) only partially explains the increased morbidity and mortality from coronary heart disease (CHD). Other factors must therefore be considered in this group of patients. Triglyceride concentrations, particularly post-prandial levels, may be important. Diabetic subjects have increased very-low-density-lipoprotein (VLDL), increased intermediate-density-lipoprotein (IDL) and low high-density-lipoprotein (HDL) concentrations, and differences in lipoprotein composition may partly explain increased atherogenesis. Although LDL levels of diabetic patients are not different from those of control subjects. LDL particles are potentially atherogenic as they are smaller, more dense and prone to oxidative modification. NIDDM subjects also have altered apolipoprotein concentrations, including increased apoB, apoC-III, and decreased apoA-I; in addition, apoE-2 may be over-represented in diabetic populations. Thus, apart from the traditional risk factors, there are several lipoprotein compositional abnormalities that may contribute to the increased prevalence of CHD in diabetes.

In 1987, as part of the World Health Organization's Inter-Health Programme, we carried out a noncommunicable diseases survey in six rural villages in Tanzania. Each women completed a questionnaire, part of which was concerned with obstetric history, and underwent a physical examination. Our findings of unacceptably high perinatal mortality and abortion rates are described below.Within the framework of the World Health Organization's [WHO] Inter- Health Program, a noncommunicable diseases survey was carried out in 6 villages located in 2 regions of Tanzania in 1987. A total of 3565 women were queried about their viable pregnancies, still-births, perinatal deaths (death within the first week of life), and abortions (loss of conceptus within 28 weeks of gestation). The overall abortion rate was 119/1000 pregnancies (range of 89/1000-170/1000 pregnancies). In the Kilimanjaro region, the abortion rate was 97/1000 compared to 145/1000 in the Morogoro region. The overall perinatal mortality rate was 73/1000 births (range of 49/1000-124/1000 births). In the Kilimanjaro region, the perinatal mortality rate was 58/1000 births vs. 91/1000 in the Morogoro region. The overall abortion rate for the 6 villages was 12% of pregnancies, while the range culled from the literature was 10-25%. The rate of miscarriage before 20 weeks of gestation has been estimated at 36-43%. In the Morogoro region, miscarriage rates were highest in the older age groups, which is indicative of improved health care. The difference in perinatal mortality rates between the 2 regions may be attributable to the relatively higher level of development of the Kilimanjaro region. This difference in socioeconomic development was also reflected in the infant mortality rate: 70/1000 births in the Kilimanjaro compared to 140/1000 in the Morogoro region. Other studies in east Africa showed perinatal mortality rates of 46/1000 births in Kenya's Machakos region during 1975-78, and a rate of 124/1000 births in Tanzania near the Morogoro region during 1984-85. Community-based studies focusing on women's health and care during pregnancy and childbirth are needed to investigate the causes of perinatal deaths, because hospital-based studies give a disjointed picture of both rates and the causes.

A study of the prevalence of schistosomiasis was carried out in 253 school children in Melela, Tanzania, one year after a single dose of praziquantel, 40 mg/kg body weight. The cure rate was 90.4%. However the new incidence estimate was 21.2% in children who were initially negative. The use of reagent strips in urine tests for the detection of urinary schistosomiasis was also evaluated. The presence of blood in the urine was highly sensitive (96%) and specific (87%). The sensitivity of proteinuria was less, but it was highly specific (94%). This study shows that chemotherapy will have to be combined with other measures to achieve lasting benefits and raises the question as to how often the population should be treated. It also confirms the value of urine test strips as an indirect diagnostic test for urinary schistosomiasis in an endemic area. Further follow-up is necessary to make useful predictions concerning incidence and reinfection rates in the community. This will also help in deciding how often chemotherapy should be given to the population.

The relationship of 2-hour (post-75 g oral glucose) serum insulin levels with plasma glucose levels was studied in a population-based random sample comprising 2,627 Hindu Indians, 685 Muslim Indians, 1,351 Creoles (African, European, and Indian admixture), and 415 Chinese from the Indian Ocean island of Mauritius. Known diabetic subjects taking oral hypoglycemic drugs or insulin were excluded from these analyses; 64% of all diabetic subjects had usable glucose and insulin data. Both fasting and 2-hour postload insulin levels were significantly higher in women than in men, and levels in both sexes were significantly greater in Hindu and Muslim Indian subjects than in Creoles or Chinese even after controlling for differences in age, body mass index (BMI), waist to hip ratio (WHR), and plasma glucose level. Levels in Muslims were higher than those in Hindus; it was unclear whether these ethnic differences represented hereditary or unmeasured environmental factors closely associated with ethnicity. All four ethnic groups demonstrated similar inverted U- or V-shaped curves when 2-hour insulin was plotted against either basal or 2-hour glucose. Both quadratic (U) and two-piece (V) regression models improved over linear models for 2-hour insulin versus either fasting or 2-hour glucose in all ethnic groups, although in statistical terms they were good models only for the 2-hour glucose comparison. The two-piece models were associated with modest increases in R2 compared with the quadratic models, but it was not possible to precisely determine optimal turning points with either model. However, in all ethnic groups, 2-hour insulin levels decreased above glucose levels of 7.1 to 7.8 (fasting) and 11.3 to 13.5 mmol/L (2-hour) in quadratic models, and 7.5 to 9.5 (fasting) and 8.5 to 10.5 mmol/L (2-hour) in two-piece models. The shape and point of inflection of the quadratic and two-piece curves were influenced little by gender, obesity, fat distribution, and physical activity. These results are in accord with those observed in cross-sectional and longitudinal studies in other ethnic groups, and support the generality of the plasma glucose levels currently used to define diabetes mellitus, which physiologically correspond with a decrease in beta-cell responsiveness to glucose. Asian Indians appear to have an ethnic propensity to hyperinsulinemia that is not explained by obesity or adverse fat distribution.

There is conflicting evidence of a link between circulating insulin concentrations and blood pressure. The aim of this study, therefore, was to investigate this relationship in a representative population sample of different ethnic groups.A cross-sectional study was performed, using a random cluster sample of 5080 adults aged 25-74 years from the Indian Ocean island of Mauritius.Glucose tolerance was assessed with a 75-g oral glucose tolerance test. Hypertension and glucose tolerance were classified using World Health Organization criteria. Subjects known to have diabetes before the survey were excluded from these analyses, leaving 3104 Asian Indian, 1256 African-origin Creole and 384 Chinese subjects eligible for this study.There were statistically significant linear correlations between fasting and 2-h insulin and mean blood pressure in all sex-ethnic subgroups, except for Chinese men. However, after controlling for age and the body mass index, partial correlations were much reduced and remained significant only for fasting insulin in Indian men (r = 0.07, P0.01) and women (r = 0.09, P0.001) and 2-h insulin in Chinese women (r = 0.15, P0.05). After controlling for age and body mass index, mean fasting and 2-h insulin concentrations were not significantly different between hypertensive and normotensive subjects in any of 36 possible subgroups defined by ethnic group, gender or glucose tolerance (normal, impaired glucose tolerance or newly diagnosed diabetes). In both men and women with normal glucose tolerance (ethnic groups combined) there was no evidence of increasing systolic or diastolic blood pressure across quartiles of either fasting or 2-h serum insulin within any quartile of the body mass index. By contrast, at any given level of serum insulin there was a clear association between an increasing body mass index and increasing blood pressure, suggesting strongly that insulin is not the final pathway by which obesity influences blood pressure.This study of three ethnic groups does not support the hypothesis that the circulating insulin concentration is an independent determinant of blood pressure in populations.

The glucoregulatory role of insulin in adult subjects is undisputed. However, less is known about the secretion of insulin and its actions in the neonatal period, either for healthy subjects, or for those at risk of disordered blood glucose homoeostasis. The relationships between blood glucose and plasma immunoreactive insulin concentrations were therefore examined in 52 healthy children (aged 1 month-10 years), 67 appropriate birth weight for gestational age (AGA) term infants, and 39 AGA preterm neonates. In children and AGA neonates, plasma immunoreactive insulin concentration was positively related to blood glucose concentration. However, although both groups of neonates had significantly lower blood glucose concentrations than children, plasma immunoreactive insulin concentrations were significantly higher in both term and preterm neonates, when compared with children. The variation in plasma immunoreactive insulin concentrations was greater for neonates than for children. These data suggest, that compared with older subjects, plasma immunoreactive insulin concentrations are high in newborn babies and that neonatal pancreatic insulin secretion is less closely linked to circulating blood glucose concentrations. There are important implications for the interpretation of studies in hypoglycaemic and hyperglycaemic neonates.

We have compared the effects of subcutaneously injected human proinsulin, insulin zinc suspension and inactive diluent (control) on glucose turnover, intermediary carbohydrate and lipid metabolism in non-insulin-dependent diabetic man. Six weight-matched (24.8 +/- 1.6 kg M-2) non-insulin-dependent diabetic subjects underwent 3 separate, randomized, 10 h isoglycemic clamps. Glucose turnover was measured using a primed continuous infusion of [6'6'2H2] glucose. Each subject received 0.35 U/kg of hormone or control made up to isovolumetric amounts. The mean blood glucose level of 7.3 +/- 0.8 mmol/l was similar at the start of each isoglycemic clamp. Incremental area under the curve proinsulin levels (1195 +/- 146 nmol/l) were about 21-fold higher, on a molar basis, than insulin (62.4 +/- 10 nmol/l). Des 64/65 split proinsulin increased in a parallel manner to intact proinsulin (r = 0.99, P0.0001) and comprised approximately 13% of the intact proinsulin concentration. Hepatic glucose production was suppressed similarly following proinsulin and insulin zinc injection. However, both proinsulin and insulin zinc had a significantly greater effect on suppression of hepatic glucose production compared to control (P = 0.01, P = 0.009, respectively). Metabolic clearance rate of glucose fell significantly during the control studies compared to insulin zinc or proinsulin injections (P0.05). Blood lactate, pyruvate and alanine concentrations were similar following control or hormone injections. However blood glycerol, 3-hydroxybutyrate and plasma-non-esterified fatty acids were suppressed significantly by proinsulin and insulin zinc compared to control injections. The conclusions were: (1) In overnight fasted hyperglycemic non-insulin-dependent subjects s.c. injections of proinsulin and insulin zinc can produce similar effects on glucose turnover, intermediary lipid and carbohydrate metabolism. (2) Similar carbohydrate intermediary metabolism profiles can be obtained following insulin zinc, proinsulin or control injections. (3) However lipolysis and ketogenesis were significantly suppressed by both hormones compared to control. (4) Subcutaneous proinsulin injection resulted in approximately 13% conversion to des 64/65 split proinsulin.

The oral glucose tolerance test is inconvenient for diabetes screening. In clinical studies a reduced serum anhydroglucitol level has proved to be a sensitive and specific test for diabetes. A new minicolumn enzymatic method which is simple and robust makes use in population screening feasible. The aim of our study was to assess the usefulness of a single measurement of anhydroglucitol to screen for diabetes and impaired glucose tolerance. Assays were performed on samples taken from 227 Mauritian Chinese subjects at the time of glucose tolerance testing. Subjects had normal glucose tolerance (n = 82), impaired glucose tolerance (n = 76), newly diagnosed diabetes (n = 38), and known diabetes (n = 31). Anhydroglucitol concentrations (mean +/- SD) were similar for normal and impaired glucose tolerance subjects (23.7 +/- 8.2 vs 23.4 +/- 8.6 mg l-1). Although the differences between normal and newly diagnosed diabetes (15.0 +/- 11.0 mg l-1) and known diabetic subjects (11.8 +/- 10.6 mg l-1) were significant (p0.001), diagnostic sensitivity and specificity were poor. We conclude that measurement of serum anhydroglucitol is not suitable for screening for the diagnoses of impaired glucose tolerance and diabetes.

We assessed the HemoCue system for measuring glucose in 5 microL of whole blood. A glucose dehydrogenase-based reaction is used with dried reagents contained in disposable microcuvettes, which are filled with blood by capillary action. Automated hexokinase and YSI 23AM glucose analyzer methods were used for comparison. Overall imprecision (CV) was better than 4.5%, with no significant differences in results between three different HemoCue photometers and four batches of microcuvettes. Regression slopes (+/- SE) were 0.947 (0.011) with the YSI and 0.966 (0.015) with the hexokinase method. Analytical recovery of added glucose was 101-106%, and the system functioned with hematocrits up to 0.65. Bilirubin up to 453 mumol/L did not interfere, but high concentrations of endogenous (greater than 3 mmol/L) and exogenous triglycerides gave positive interference. The system proved stable and robust under a wide range of storage and handling conditions; performance was impaired only at high ambient temperature (37 degrees C). We conclude that the HemoCue system should prove useful for glucose measurement; further testing outside the laboratory is warranted.

The association between plasma fibrinogen levels, fibrinogen genotype, and the development of macrovascular disease was studied in 100 patients with non-insulin dependent diabetes mellitus (NIDDM). The mean plasma fibrinogen levels in patients with macrovascular disease was higher than those without, although the difference was not statistically significant (3.67 g l-1, and 3.43 g l-1, respectively). The frequency of the rare allele of the fibrinogen gene DNA polymorphism detected with the restriction enzyme Bc1I was slightly higher in the group of patients with disease, but the difference was not statistically significant (0.20 vs 0.16). The frequency of the TaqI polymorphism rare allele was the same in both groups (0.30 vs 0.31). However, the Bc1I polymorphism was strongly associated with plasma fibrinogen levels, with those patients heterozygous for the rare allele having mean levels 16 per cent higher than those lacking the allele (3.81 g l-1 vs 3.28 g l-1, p0.05). This data demonstrates that variation at the fibrinogen locus is involved in determining fibrinogen levels in patients with NIDDM, and suggests the possibility that fibrinogen genotype and plasma fibrinogen levels could be one of the factors making a small contribution to the development of macrovascular disease in diabetic patients.