Prevalence of the Metabolic Syndrome Among Adult New Zealanders of Polynesian and European Descent

RESEARCH DESIGN AND METHODS

Residents from a household census in South Auckland were randomly selected using a sampling framework stratified by age, sex, and ethnic group as previously described (2). At each house, questionnaires were completed; weight, height, and waist were measured; and a random venous sample was taken for glucose measurement (into fluoride; spun, chilled, and assayed within 3 h). Those without known diabetes and either a random glucose ≥6.5 mmol/l within 2 h of a meal or ≥6 mmol/l ≥2 h after a meal (defined as “screen positives”), as well as a random control group (28% of other subjects; defined as “screen negatives”) were invited for a 75-g oral glucose tolerance test (OGTT). At the OGTT, fasting samples were also taken for lipids (total, HDL, and LDL cholesterol and triglycerides) and measured on a Hitachi 747 analyzer (Hitachi, Tokyo, Japan) (interassay coefficient of variation 2.2, 4.8, and 3.3%, respectively, except LDL cholesterol, which was calculated using the Friedewald equation). Resting blood pressure was taken as the mean of two measurements using a standard mercury sphygmomanometer (with large cuff where necessary) following a standardized approach and before the 2-h glucose sampling.

The ATP III criteria (2) for the metabolic syndrome were considered to have been met when three or more of the following factors were present: waist circumference >102 cm for men or >88 cm for women, elevated blood pressure (treated hypertension or systolic blood pressure ≥130 mmHg and/or diastolic blood pressure ≥85 mmHg as mean of two readings), elevated triglycerides (≥1.7 mmol/l), decreased HDL (<1.04 mmol/l for men or <1.29 mmol/l for women), elevated fasting glucose (≥6.1 mmol/l), or diabetes. The study was approved by the University of Auckland Ethics Committee, and subjects gave informed consent.

Randomization and statistics were undertaken using SPSS for Windows (version 12.0; SPSS, Chicago IL). Comparisons of proportions were made using the χ² test. Calculations of the prevalence of each constituent of the metabolic syndrome by ATP III criteria and overall were made by direct standardization to the proportion of subjects who were screen negative or positive with Excel 2000 (Microsoft) as follows: (screen positive with characteristic at OGTT) × (% screened who were screen positive)/screen positive attending OGTT + (screen negative with characteristic at OGTT) × (% screened who were screen negative)/screen negative attending OGTT.

RESULTS

As previously reported, the initial household survey had a 91% response with 2,737 interviewed, of whom 314 had known diabetes. Of the 2,423 invited to participate without known diabetes, 1,562 (64.5%) were screened in their home. Subsequent attendance at OGTT was 68.4% (327 of 478) among screen-positive and 67.2% (207 of 308 invited) among screen-negative subjects. Mean age was 50 ± 5 (Europeans), 49 ± 5 (Maori), and 49 ± 5 years (Pacific) (P = 0.029) in the 40- to 59-year age-group and 68 ± 6, 66 ± 5, and 67 ± 5 years, respectively, in the 60- to 79-year age-group (P < 0.001).

The prevalence of known diabetes, obesity, and low HDL was significantly higher among Polynesians in both age-groups (Table 1). Maori, but not Pacific, subjects consistently had a higher blood pressure, although this difference was not significant among women aged 60–79 years. The prevalence of hypertriglyceridemia was highest among Maori and Pacific people, except among women aged 40–59 years. The presence of an elevated fasting glucose or new diabetes diagnosis was significantly more likely among Maori and Pacific women aged 40–59 years and nonsignificantly higher among men of these ethnic groups.

Among those aged 40–59 years, a very small proportion of Maori and Pacific subjects showed no components of the metabolic syndrome and a much higher proportion (approximately half) had three or more components, making them eligible for metabolic syndrome diagnosis. Among those aged 60–79 years, there was no significant difference in the overall distribution of the components or prevalence of the metabolic syndrome among women (P = 0.350). The prevalence of the metabolic syndrome was higher among men (P = 0.005).

CONCLUSIONS

Among Maori and Pacific peoples aged ≥40 years, the total prevalence of the metabolic syndrome and known diabetes was >50%, with ∼80% having at least two components of the metabolic syndrome (also associated with a significantly increased risk of cardiovascular disease [8]). The prevalence of the metabolic syndrome by ATP III criteria among men of European descent aged ≥40 years has been reported to range from 12% across the eight DECODE (Diabetes Epidemiology: Collaborative Analysis of Diagnostic Criteria in Europe) studies (8), to 28.8% (35–75 years) in Canada (9), and to 30–40% in the U.S. (10). These New Zealand data are clearly most comparable to those in North America. Among women of European descent in Canada aged 35–75 years, the prevalence of the metabolic syndrome was 14.3% (9), and in Europe, the prevalence increased from 5.2% among those aged 40–49 years to 16.8% among those aged ≥70 years (8). In the U.S., the prevalence of the metabolic syndrome increased from ∼19% among women aged 40–49 years to a peak of ∼45% in those aged 70–79 years (10). There have been no reports of the prevalence of the metabolic syndrome among Polynesians elsewhere, but the age-specific prevalences among Maori and Pacific people were comparable to Mexican Americans (10,11), Arab Americans (12), and Native Americans (9,13) but higher than those among South Asians (9), East Asians (9,14,15), and African Americans (10).

These findings are limited by the small numbers in each stratum and the stratified nature of the attendance at OGTT, and this probably explains the lack of statistical significance for some components (e.g., new diabetes). The criteria for obesity are also an issue among Polynesians (16).