Prevalence of Prediabetes Based on Fasting Plasma Glucose and Glycosylated Hemoglobin in an At-Risk Mexican Population

Introduction

Prediabetes or a prediabetic state is defined as a condition characterized by glucose levels above normal but below the threshold of diabetes. ¹ This term is used for patients with impaired fasting glucose and/or impaired glucose tolerance and indicates an increased risk for type 2 diabetes (T2DM) and cardiovascular disease. ² Statistics show that around 5%–10% of people with prediabetes become diabetic annually and that up to 70% will develop T2DM during their lifetime. ^2,3 In Mexico, Coronado et al. ⁴ reported a 5.9% annual conversion rate. Regarding the pathophysiology, the transition from normoglycemia to prediabetes begins with insulin resistance, a condition usually triggered by obesity. The status of resistance results in prolonged compensatory hyperinsulinemia followed by a period of stable adaptation in which beta-cell insulin secretion decreases but maintains normal blood glucose; once it is unable to compensate for insulin resistance, the glucose level increases. ⁵

It is important to mention that in the prediabetes stage, complications such as nephropathy, neuropathy, retinopathy, and macrovascular disease begin to develop. However, the diagnosis of prediabetes offers patients the opportunity to prevent the development of T2DM since the former is reversible with pharmacological and nonpharmacological treatment, which can reduce the risk of conversion by up to 70%. ¹ Globally, fasting plasma glucose (FPG) levels have increased in both developed and developing countries with a subsequent increase in the incidence of prediabetes and T2DM. ⁶ In the United States, the estimated prevalence of prediabetes is 37%, which increases with age. ^7,8 In England, it is reportedly 35.3% ⁹ ; in Mexico, it is 43.2%. ¹⁰ Prediabetes screening is commonly performed in at-risk individuals defined as sedentary, overweight–obese, with personal history of dyslipidemia, hypertension, cardiovascular disease, and a first-degree family history of T2DM. ²

A common test is FPG with glucose levels of 100–125 mg/dL and/or a 2-hr oral glucose tolerance test with a postload plasma glucose of 140–199 mg/dL. In recent years, glycosylated hemoglobin (HbA1c) has been recommended because of its ability to predict T2DM, which expresses an average glycemia (fasting and postprandial) of the previous 3 months without the need to fast. ^{2,11 –13} It is notable that abnormal HbA1c levels, even in a nondiabetic population, increase one's cardiovascular risk. ¹ Some studies that aimed to measure the prevalence of prediabetes found a lower frequency of prediabetes using the HbA1c criterion than using FPG. James et al. documented a prevalence of 14.2% based on HbA1c and 26.2% based on FPG, ¹⁵ while Bullard et al. reported 19.3% versus 27.5%, respectively. ⁸ It is possible that the combined use of these two tests can better detect individuals at increased risk of T2DM than the use of either alone, ¹² since positivity to both tests represents a worse metabolic profile. ¹⁶

There are no published studies in Latin America that consider the prevalence of prediabetes according to the criterion of HbA1c in addition to FPG. The relevance lies in pointing out the potential reduction in the incidence of T2DM given the reversibility of prediabetes. For example, in the northern states of Mexico, an increase in the prevalence of T2DM was reported from 6.4% to 11.4% between 2006 and 2012 ^17,18 ; certainly, the disease could have been delayed or prevented in a large percentage of that population if the prediabetes stage had been diagnosed and treated. The aim of this study was to determine the prevalence of prediabetes using FPG and/or HbA1c in a Mexican population at risk for T2DM and cardiovascular disease.

Materials and Methods

This cross-sectional study was conducted between 2012 and 2013 and included 384 patient users of a primary care center operated by the Instituto Mexicano del Seguro Social. This center has the largest number of beneficiaries of the metropolitan area of Monterrey, Mexico, a northern region of the country. We included only individuals at least 20 years of age without a known diagnosis of prediabetes or T2DM and with at least one risk factor for T2DM and cardiovascular disease. Those with anemia or who were treated with corticosteroids were excluded, as were pregnant women. The sample size was calculated based on a 50% expected prevalence, 95% confidence level, and 5% bound error. All participants signed informed consent. The protocol was submitted to and approved by the institute's research and ethics committee.

A previously trained family medicine resident interviewed participants on the following T2DM and cardiovascular disease risk factors: first-degree relative diagnosed with T2DM and personal history of hypertension, ischemic heart disease or dyslipidemia, and physical inactivity (lack of regular physical exercise). Following the survey, the participants were weighed while dressed in lightweight clothing and not wearing shoes, while their heights were measured using a mechanical scale with a stadiometer. Nutritional status was categorized on the basis of each participant's body mass index (BMI) value (normal weight, BMI 18.5–24.9 kg/m²; overweight, BMI 25.0–29.9 kg/m²; and obese, BMI ≥30.0 kg/m²). Individuals with at least one risk factor were scheduled the next morning in the laboratory. FPG and HbA1c were measured in venous blood samples drawn from all participants after fasting for 8–12 hr. FPG determination was performed using the glucose oxidase method, while HbA1c was measured using a turbidimetric inhibition immunoassay. The following criteria were considered positive for prediabetes: (1) FPG 100–125 mg/dL; and/or (2) HbA1c 5.7%–6.4%. ² Those patients with abnormal FPG and/or HbA1c values were followed in order to confirm the diagnosis; the participant was referred to their physician's office for monitoring and appropriate treatment, if any lab result was abnormal.

Central tendency and dispersion were determined for quantitative variables and proportions were determined for qualitative variables. The point prevalence of prediabetes and 95% confidence intervals (CI) were estimated overall and stratified by age (<50 years, ≥50 years), sex, and nutritional status. Comparisons of FPG and HbA1c were performed using a Mann-Whitney test since they showed a non-normal distribution, while qualitative variables were compared using a chi-squared test. P values < 0.05 were considered significant.

Results

The study population consisted of 252 women (65.6%) and 132 men (34.4%) with a mean age of 47.5 ± 12 years (range, 20–65 years); mean BMI was 31.2 ± 6.0 kg/m² and physical inactivity was the predominant risk factor (Table 1). Mean FPG and HbA1c were 108.6 ± 20.6 mg/dL and 5.8% ± 0.7%, respectively; both were higher in individuals ≥50 years and in those with overweight–obesity (Table 2).

Discussion

The prevalence of prediabetes based on FPG and/or HbA1c was estimated in a study population of young adult primary care users with at least one risk factor for T2DM or cardiovascular disease. It is important to note that FPG was an average of 9 mg/dL higher than the upper normal limit established by the American Diabetes Association ² and even higher than the limit of 97.5 ± 30 mg/dL in primary care users reported in 2008 in a Mexican study. ¹⁰ The most important findings are discussed below.

Six of 10 participants were diagnosed with prediabetes using FPG alone, a rate higher than that reported in North America (27.5%) ⁸ and in Mexico (24.6%). ¹⁰ As expected, the prevalence of prediabetes using HbA1c was lower than that using FPG ^8,15,16 ; however, it was close to 50%, higher than the rates reported in the United States (19.3%) ⁸ and England (35.3%). ⁹ Further,> 30% of the population tested positive on both FPG and HbA1c in contrast to 9.6% of the population tested in the United States. ⁸ The different prevalence of prediabetes observed between FPG alone, HbA1c alone and FPG-HbA1c may reflect different states of insulin resistance progression. During the impaired fasting glucose stage, patients experience a decline of the first-phase insulin secretion and when the level of HbA1c is increased, patients experience a decline of both early and late phase insulin secretion. ¹⁶ For that reason, HbA1c alone or combined with FPG represent greater metabolic impairment and consequently, higher risk for developing diabetes. HbA1c reflects exposure to the basal and postprandial hyperglycemia for the preceding 8–12 weeks, which makes it an imminent marker of insulin resistance and, therefore, greater metabolic impairment. ¹⁹ HbA1c levels of 5.5%–6.4% are known to be strongly associated with the incidence of diabetes the next 5 years, ¹¹ which implies the need to recognize a subpopulation with a higher care priority given the intense risk of conversion to T2DM. ¹⁹

It was expected that we would observe a higher prevalence using FPG or HbA1c rather than both. We identified seven of 10 participants with prediabetes using one test or the other; the overweight–obesity and ≥50 years of age groups were the most commonly affected. This global frequency far exceeded that reported in the USA of 37%. ^7,8 This figure is alarming assuming a 5.9% annual incidence of conversion to T2DM. ⁴ It is urgent that we strengthen health programs aimed at reversing prediabetes to delay or prevent the development of T2DM.

Limitations

A population-based study would have been desirable. This investigation demonstrated the need to identify prediabetes in all patients who visit the clinic for reasons other than prediabetes and T2DM. However, caution is needed to generalize these results to suburban or rural primary care users. These participants were identified as being at risk for having at least one risk factor for T2DM or cardiovascular disease, with support in screening targeted at priority groups according to the policy of any health institution with limited resources. Importantly, the general Mexican population may qualify as being at risk for prediabetes by the alarming rates of overweight and obesity. The latest National Health and Nutrition Survey in 2012 revealed that 73% of adult men and 69% of adult women were overweight or obese. The high prevalence of physical inactivity in the present study was not surprising given the national statistics that reported sedentary and inactive activities almost 16 hr each day (sleeping, sitting in front of a screen, inactive transportation). ¹⁸

Conclusions

The prevalence of prediabetes in a Mexican population at high risk of cardiovascular disease and/or T2DM was 50%–70% depending on the positivity criteria of HbA1c alone, FPG alone, or both. One-third of the 10 patients identified with prediabetes tested positive on both FPG and HbA1c. Prediabetes is definitely a public health problem, and future studies are needed to determine the effectiveness and efficiency of pragmatic strategies to reverse the status of prediabetes and therefore reduce the incidence of T2DM.