Characterization of Adults with Initial Diagnosis of Type 2 Diabetes Mellitus in Peru

Introduction

Both the prevalence and incidence of type 2 diabetes mellitus (T2D) is on the raise, with an estimated 425 million adults (ages 20 to 79 years) living with this medical condition, and with the bulk of the burden affecting low- and middle-income countries [1, 2]. In Peru, screening studies have estimated that the prevalence of diabetes mellitus is 5,1% among those older than 35 years, and 7% among those older than 25 years [3]. Multiple factors have been associated with higher risk of T2D, including living in the northern provinces, and having migrated from a rural to an urban area [4]. In contrast, some studies have found an inverse association between living in the highlands and the incidence of T2D [5, 6]. However, there is scarce data about the main clinical and laboratorial characteristics of subjects with an initial diagnosis of T2D. A study in the United States analyzed demographic characteristics of patients with initial diagnosis of T2D and reported that younger patients (ages 18 to 45 years) tended to be female, have higher body mass index (BMI), have poorer glycemic control and higher prevalence of systolic hypertension, in comparison to those who were initially diagnosed at age > 45 years [7].

Understanding the characteristics of subjects with initial diagnosis of T2D would be valuable to define strategies for earlier diagnosis and timely therapeutic interventions in order to prevent the progression of the disease and to optimize long term outcomes. Even preventive strategies could be informed should factors could be modified in order to delay the onset of the disease. The main objective of this study was to describe the main clinical and laboratorial characteristics (age, sex, BMI, abdominal waist circumference, arterial blood pressure, glucose levels, glycosylated hemoglobin, lipid profile, and presence of fatty liver) of adults who received an initial diagnosis of T2D in several regions of Peru. Given that the sites were distributed across multiple regions of the country, it was possible to explore potential differences across regions including those located at high altitude.

Subjects

Patients with the following criteria were included: a) age > 20 years, b) ambulatory, c) having received an initial diagnosis of T2D. The diagnosis of T2D was defined with the following parameters: fasting glucose ≥ 126 mg/dl, or glucose tolerance test >200 mg/dl at 2 hours, or random glucose ≥ 200 mg/dl along with classical T2D symptoms. A glycosylated hemoglobin (HbA1c) ≥ 6.5% was considered as a diagnostic element when the glycemia was borderline, that is, T2D was not diagnosed solely on the bases of elevated HbA1c. In order to reduce the likelihood of including adults with type 1 diabetes, in patients with ages between 20 and 30 years, at least one of the following additional criteria was required: body mass index (BMI) > 30, increased waist circumference, first degree family history of T2D, or acanthosis nigricans.

Subjects with use of drugs known to alter glucose metabolism (such as corticosteroids, loop diuretics, and cancer chemotherapy) were excluded. If the patient had previously used metformin for treatment of a medical condition other than diabetes, metformin would have to be discontinued for at least two weeks prior to the diagnosis of T2D. Exclusion criteria included pregnancy and having renal, hepatic, respiratory, or cardiac (NYHA class 3-4) insufficiency.

Methods

This was a descriptive case series study performed at 26 centers, primarily ambulatory, in multiple provinces of Peru. Centers located at high altitude (>1894 meters above sea level) were included. Participating investigators, largely endocrinologists, were invited to collect medical record data from adults who had been given an initial diagnosis of T2D. Data collection was performed between July 2016 and November 2017.

Results

Information was collected from 581 patients who had received an initial diagnosis of T2D. Females composed 48.9% of the participants. The mean age was 53.2 years (SD +/- 12.8 years). Early diagnosis of T2D (before age 40 years) was found in 18.3% of the patients. Most (54.7%) received their initial diagnosis of T2D between the ages of 41 and 60, which was considered as middle age of diagnosis, for comparison. The remaining (27%), received the initial diagnosis of T2D after 60 years of age and were considered as late onset T2D.

Most (68%) of the participants were found to have increased waist circumference according to different definitions. When the cut-off points of ≥ 94 cm for men or ≥ 90 cm for women were considered, an increase in waist circumference was found in 79.29% individuals, with no significant difference between men and women. However, when the cut-off points of ≥ 102 cm for men and ≥ 88 cm for women were used, an increase in the waist circumference was found in 68.9%, with men having 73% lower risk of having a high waist circumference in comparison to women (OR = 0.27, 95% CI 0.17 to 0.41).

Regarding the results of fasting blood glucose, the average was 249.4 ± 105.7 mg/dl and 61.7% had values above 200 mg/dl. In addition, it was observed that individuals with glycemia> 200 were more likely to be male (OR: 2.83; 95% CI 1.83 to 4.38). Regarding HbA1c, an average of 8.9 ± 2.4% was observed. Among those with HbA1c > 9 % (n = 172, 41.1%) being male had a higher odds ratio (OR: 2.21 95% CI 1.42 to 3.42). More details are shown in table 1.

Discussion

This study describes the main clinical and laboratorial characteristics of newly diagnosed patients with T2D at different regions of Peru, including centers located at high altitude. Most patients were middle aged with 84.4% having excess weight; the mean blood glucose and HbA1c were 249.4 mg/dl and 8.9%, respectively.

As there is no national validation of waist circumference cut off, we evaluated different definitions such as that of Aschner et. al. that was developed for Latin American population (≥ 94 cm men or ≥ 90 cm women) and a definition made for the US population (≥102 cm men or ≥88 cm women) [8, 9]. Using both, an increased waist circumference was observed in more than 68% of the individuals. This high percentage in the newly diagnosed diabetic population coincides with other study published in Peru that found a frequency of abdominal obesity of 65.6% [10].

In 39.2% of the participants the BMI at diagnosis fell in the overweight category, while 45.1% fell in the obese category. Different studies in the Peruvian general population have found percentages of overweight and obesity ranging from 40% to 62% [11]. In a study published in Peru, the frequency of non-morbid obesity (BMI 30-39) and morbid obesity (BMI> 40) was estimated through different national surveys from 1975 to 2013. The authors found that non- morbid obesity increased from 8.5% to 18.5%, while morbid obesity had only a 1% increase in the same period [12]. It is important to highlight that 69.9% of this sample of diabetic patients fell in the range of overweight and mild obesity, while 14.5% had a BMI greater than 35.

Centers in the north and east had a higher percentage of overweight, while those in Lima had higher percentage of obese subjects. However, there were no significant differences in waist circumference between individuals according to geographical location. The highest values of glycemia and glycosylated hemoglobin were observed in the North and Lima. This observation is relevant to prioritize interventions to the locations showing highest risk. It has been reported that subjects living at high altitude have lower prevalence of T2D, lower frequency of obesity and shorter waist circumference [13, 14, 15]. In our study, the phenotype of the diabetics living at high altitude was similar to that of those living at sea level, except for levels of total cholesterol and LDL-c that were lower at high altitude. Such results may be due to differences in dietary habits.

In 56.7% of patients with newly diagnosed T2D who had abdominal ultrasound, findings were compatible with hepatic steatosis. In addition, there was higher frequency of fatty liver in diabetics from high altitude, while there was no difference by age at diabetes debut. There was lower percentage of fatty liver in subjects from the north of the country, in contrast to the higher rates of other risk factors observed among these patients [16, 17, 18].

Recent studies have shown that the characteristics of patients with T2D differ according to the age at which the diagnosis is made [19, 20]. Our results show that the majority of patients diagnosed at an advanced age had a previous diagnosis of prediabetes (self-report), and lower frequency of first degree family history of diabetes compared to the other groups. This coincides with previous studies that have shown that patients diagnosed at an early age are more likely to have a family history of diabetes, which may reflect their genetic burden of T2D [21]. We found no significant differences in HDL-c, LDL-c or triglycerides. This coincides with a study in China that also found no difference in these values [22]. However, it is important to note that elevated total cholesterol was more frequent in the early age group compared to the advanced age group. In addition, an important finding is that elevated glycosylated hemoglobin (greater than 10%) as well as fasting glucose levels greater than 200 mg / dL was observed more frequently in the early diagnosis group compared to those of diagnosed at middle or advanced age. This finding suggests that the metabolic involvement of these patients is more severe than in the other groups and represents an important information to guide the prioritization of interventions.

One of the limitations of the study is inherent to its design as a case series study without a control group (not diabetes). In addition to the descriptive statistics provided, different associations were explored. However, such associations could only generate hypothesis for future studies. Another limitation is the representativeness of the sample, given that convenience sampling was used. Only subjects seeking medical attention could be included. Furthermore, hospitalized subjects were not included. Ultrasound findings suggestive of liver steatosis could not be confirmed by other methods.

As a strength, this is one of the first studies focusing on newly diagnosed patients with T2D in Peru. Laboratory tests and imaging studies were included in our study, in contrast to prior epidemiological studies. The wide distribution of sites at multiple regions is a strength that provides a broad representation of the country. Of particular value was the possibility to explore associations of different variables with location at high altitude. The analysis of different characteristics allows for identification of populations at higher risk and may inform strategies for the prevention and management of this prevalent disease.

Conclusion

This is one of the first studies describing the characteristics of adults with an initial diagnosis type 2 diabetes mellitus in Peru. Subjects from Lima and the north of the country were found to have higher metabolic risk, as well as those diagnosed at an early age. These findings are relevant to inform interventions for prevention and management of this prevalent disease in Peru.

Funding Source

Honorarium for statistical analysis by Dr. Alvaro Whittembury was paid via a grant from Novo Nordisk Peru SAC. Otherwise, this research did not receive any other specific grant from funding agencies in the public, commercial, or not-for-profit sectors.