Determining A1C Normal Values in Rhesus Macaques (Macaca mulatta)

Introduction

According to the American Diabetes Association, 1.4 million people will be diagnosed with diabetes this year. Diabetes is a worldwide health issue affecting ∼1 in 10 individuals between 20 and 79 years of age.¹ It is expected that the number of people suffering from diabetes will increase due to factors such as population growth, urbanization, and overall increased incidence of obesity and lack of exercise.² Anticipated incidence of diabetes is expected to increase in future years, and by the year 2030 it is expected that an additional half billion people will be at risk for the development of diabetes, and in individuals >64 years of age, the number of people affected is expected to surpass 48 million in developed countries.^2–4 Approximately 50% of the people with diabetes remain undiagnosed.⁵

Diabetes is a metabolic disorder due to hyperglycemia resulting from insulin resistance or insulin secretory defects.^5–7 The American Diabetes Association classifies diabetes into 4 categories, type 1 diabetes mellitus, type 2 diabetes mellitus (T2DM), gestational diabetes mellitus, and other causes of diabetes. Type 1 diabetes mellitus is mainly due to an autoimmune destruction of pancreatic β cells causing an absolute deficiency in insulin, resulting in insulin dependence. Type 1 diabetes mellitus is only responsible for ∼5% to 10% of the cases of diabetes. T2DM, which accounts for ∼90% to 95% of the cases, is noninsulin dependent and due to insulin resistance and/or impaired insulin secretion rather than an absolute insulin deficiency.^1,4–12 Gestational diabetes and diabetes due to other causes also contribute to a small number of cases.^7,9,13

T2DM incidence continues to increase in developed countries and has been identified as a global health problem. With obesity being the most important risk factor, unhealthy lifestyle factors, such as lack of exercise, smoking, alcohol consumption, Western-style diets, and increased consumption of processed meats and sugar-sweetened beverages contribute to development of T2DM.^5,11,14,15 T2DM is increasing in children, adolescents, and young adults primarily due to excessive weight gain or obesity. These newly diagnosed young individuals have an accelerated development of cardiovascular disease and reduction in life expectancy.^1,15–17 While general obesity is a major contributing factor to the development of T2DM, the distribution of adipose tissue, predominately intra-abdominal and intrahepatic lipid content, appears to be more important.^1,11,13–15 Intra-abdominal visceral fat can disrupt insulin metabolism and is usually signaled by increased BMI, waist/hip ratio, waist/height ratio, and waist circumference.¹⁴

Complications from T2DM include hypertension, hyperlipidemia, strokes, cerebral and peripheral vascular disease, retinopathy, nephropathy, neuropathy, lipoprotein disorders, and cardiovascular disease, which is the primary cause of death in both prediabetic patients and patients with T2DM.^5–7,13,18 Twenty percent to 30% of diabetic patients develop complications prior to diagnosis with T2DM.⁵

While T2DM continues to grow as a major health issue, it has been demonstrated that increases in physical activity, caloric restriction, and weight loss can reduce the risk of T2DM as well as secondary complications by lowering blood glucose and insulin levels and increasing insulin sensitivity in both humans and NHPs.^{1,5,7,8,11,13,15,19}

Although fasting blood glucose, urine glucose concentrations, oral and intravenous glucose tolerance tests, ketone concentrations, and fasting plasma insulin concentrations are all still important tools in the diagnosis of diabetes in the United States, the American Diabetes Association now advocates incorporating glycated hemoglobin (A1C) in the diagnosis for its highly specific results and convenient method of screening for diabetes.^15,20–23 A1C is a measure of the amount of glucose bound to hemoglobin proteins.^3,24–26 A1C has a higher repeatability than fasted blood glucose.²¹ Unlike fasted blood glucose measurements, A1C can be collected at any time and is independent of the length of fast or content of a previous meal.^{3,12,18,21,22,24,27,28} A1C reflects average blood glucose values of 2-3 months including postprandial spikes.^{3,7,13,19,21,23,27–34} A1C is also highly correlated with blood glucose levels.^{6,12,21–23,29,34} A1C also has low intraindividual or biologic variability, relatively unaffected by acute stress or illness, and demonstrates a high predictive value at identifying undiagnosed cases of diabetes, which is estimated to be 15%-25% of all adults meeting the criteria for diabetes.^{3,12,13,18,21,22,24,28,34}

A1C is equally advantageous and widely accepted in determining chronic glycemia in monitoring diabetes as compared with blood glucose levels. It is a more complete measurement of glycemic regulation and is critical to the management of the disease. A1C levels also correlate with the risk of complications from diabetes.^{3,4,13,18,21–23,30,31,35} Point-of-care A1C kits have proven accurate on both venous and fingerstick samples.²⁰

Several species, such as cats and pigs, can naturally develop T2DM and can also develop secondary complications such as retinopathy, peripheral neuropathy, and vascular lesions. However, NHPs spontaneously develop T2DM with similar characteristics or risk features to humans, including obesity and age. Further, they are more closely related, and they exhibit clinical features similar to what is observed in humans, such as insulin resistance, dyslipidemia, and pancreatic pathology, making them excellent models for human T2DM. NHPs also exhibit a similar prediabetic period as humans.^8,9,15,24 Because NHPs are good models of T2DM and A1C levels are commonly used in human patients for diagnosis and monitoring of T2DM, it is important to develop normal values for A1C in NHP species that are used as models for human disease.^8,9,15 Currently, there are no formally accepted normal ranges for A1C in rhesus macaques. In this study, we tested 210 rhesus macaques to determine normal A1C values for healthy, nonobese adults.

Materials and Methods

Results

Our mean A1C value was 5.92% with a range of 4.4%-9.9% in all animals. Males had a mean of 6.07% with a range of 4.9%-9.9%, and females had a mean of 5.69% with a range 4.4%-6.9%. We did have 3 male outliers with an A1C of 7.5%, 8.1%, and 9.9%. The point-of-care glucometer mean for all animals was 59 mg/dL with a range of 27-95 mg/dL. The mean for the males was 58.66 mg/dL (range, 27-91 mg/dL), and the mean for the females was 59.52 mg/dL (range, 30-95 mg/dL). The blood glucose results from the chemistry panel evaluated on the Beckman AU480 analyzer had a mean of 69 mg/dL for all animals with a range of 39-111 mg/dL. Males had a mean of 67.6 mg/dL (range, 42-100 mg/dL), and females had a mean of 71.2 mg/dL (range, of 39-111 mg/dL) (Table 1).

We investigated the relationship between blood glucose levels by both point-of-care glucometer or chemistry panel and A1C and found that there was no correlation between the 2 methods of determining blood glucose values and A1C in healthy adult rhesus macaques. The correlation coefficients for A1C and blood glucose by the chemistry panel was r = 0.09, P = 0.154, and between A1C and the point-of-care glucometer was r = 0.05, P = 0.45 (Figure 1A and B). We also found that there was no correlation between A1C and weight independently (Figure 2). There was no correlation between A1C and BCS and A1C and age (Figures 3 and 4). We analyzed the data by both including and excluding the 3 most prominent outliers, and in both cases, sex was a significant factor in A1C levels with males having significantly higher A1C than females (P < 0.05) (Figure 5). When we excluded the outliers, we found that there was a significant interactive effect between sex and weight when evaluating A1C (P = 0.03) (Figure 6).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

With regard to the most prominent outliers, all were males, two 7 years of age and one 6 years of age, two with a BCS of 2.5/5, and one with a BCS of 3.5/5. The A1C values were 7.5%, 8.1%, and 9.9%, and these were the only values obtained. Our IACUC protocol was approved to collect at already scheduled sampling for screening for assignment or shipment to other licensed facilities for research, and this did not afford another opportunity to repeat their A1C determinations. The range for all blood glucose values throughout their lives for all 3 males was between 16 and 89 mg/dL. The animal that recorded a value of blood glucose value of 16 mg/dL had presented previously for a seizure prior to this study, which represented the low blood glucose value provided through clinical pathology. The fasted blood glucose values at the time of A1C sampling was 75, 71, and 74 mg/dL by the point-of-care glucometer and 79, 85, and 71 by chemistry panel through the TNPRC clinical pathology laboratory, clearly within normal range for rhesus macaques. While we do not have a definitive cause for the 3 outliers, a possibility could be simply interindividual difference in red blood cell life span or other biologic factors that may impact A1C levels independently of blood glucose levels.^35,36

Discussion

While A1C values have long been established for humans, we still do not have complete standards for NHPs, particularly rhesus macaques, which serve as a very good model of diabetes in humans. NHPs are critical as models of diabetes for intervention, therapeutic, and pathogenesis studies. Numerous species spontaneously develop diabetes, including macaques, vervets, mandrills, baboons, chimpanzees, and New World monkeys. These NHPs display clinical features similar to humans such as obesity, insulin resistance, dyslipidemia, and pancreatic pathology, making them useful animal models of human diabetes.^8,9,15

While the NHP is a valuable model for T2DM, it does have some limitations. As previously stated, NHPs can spontaneously develop T2DM; however, it can take considerable time for progression of disease, and the number of animals that spontaneously develop disease is small. There are models for development of T2DM that are diet induced by feeding high-fat or Western diets or administration of high-sugar solutions such as sucrose or fructose. This can induce metabolic syndrome and progression to T2DM, but reports vary on the success of this induction model or the duration of required feeding of high-fat or high-sugar diets.^8–10,15

There have been a few previous studies that reported A1C values for NHPs, but studies involving macaques had few numbers of animals incorporated. One study including 22 rhesus macaques demonstrated A1C concentrations for normal animals at 3.22% ± 0.55%; impaired glucose regulated animals at 4.04% ± 0.45% and T2DM animals at 4.96% ± 2.5%. Only one animal had an A1C value of >6.5%.¹⁰ Another study involving rhesus macaques examined A1C values for control, normal animals, as well as animals fed high-fat diets. Normal A1C for healthy, control rhesus macaques was between 2% and 4%, but there were only 10 animals included aged 4-10 years. Rhesus macaques with A1C levels >4.5% to 5.0% indicated a high risk of developing T2DM.⁶ One study involving cynomolgus macaques (Macaca fascicularis) found A1C levels lower than humans, with normal values of 4.4% ± 0.1% and a range of 3.0%-5.0%. That study included only 6 healthy macaques and noted that reference ranges for A1C had not yet been developed in many NHP species.²³ Another study evaluated 6 nondiabetic Celebes crested macaques (Macaca nigra) and found an average A1C value of 2.6% ± 0.2%.³⁷ One study evaluated A1C from NHPs in a zoo setting, represented by 18 species of haplorhines and 4 species of strepsirrhines. They found mean A1C values of 4.94% for haplorhines and 2.59% for the strepsirrhines. There were only minimal numbers of animals of the same species included, and only one species of macaque, the lion-tailed macaque (Macaca silenus); and the study recognized that establishment of reference ranges is still needed for NHPs.²⁴ A study examining A1C values in 81 healthy chimpanzees, evaluated both male and female, and 3 diabetic female chimpanzees with an average age of 23.9 years. The study defined a healthy chimpanzee as an animal not currently on medications and having no previous diagnosis of cardiovascular, renal, or other disease. The authors found, similar to our results, that neither age nor body weight was a significant predictor of fasted blood glucose values and A1C levels. They also found that sex was not a significant predictor of A1C levels, which differed from our findings. Further, they determined normal A1C ranges for chimpanzees, with ≤5% for healthy animals, 5.1%-5.2% for prediabetic chimpanzees, and ≥5.3% for diabetic animals.³⁸ One study involving 30 nondiabetic cynomolgus macaques with an age range of 3.5-8.5 years revealed A1C results closer to those found in our study. Specifically, the results from the point of care were 4.9%-6.4% and values were 3.9%-4.7% from a commercial laboratory. That study also evaluated A1C for diabetic animals and found that the point-of-care tests consistently yielded higher values. The point-of-care test kit used was the HbA1c test kit A1CNow⁺, the same point-of-care kit we employed, and the commercial laboratory used a Premier Hb9210 Hb A1C analyzer for determination of A1C.³⁹

The normal A1C range for healthy human adults is <5.6%. Humans with A1C levels between 5.7% and 6.4% are considered prediabetic and levels ≥6.5% are considered diabetic.^{4,6,7,13,18,20,27,32} Individuals with A1C >6% are considered at high risk for development of T2DM and should undergo routine monitoring.^7,13,21 Our mean A1C values of 5.92% in all animals, 5.69% in females, and 6.07% in males are slightly higher than previously reported in the literature for NHPs and human patients. One explanation why our A1C values were higher is that A1C levels may vary across different species of NHPs. Differences exist in A1C values among human demographics or ethnicities.^{3,17,25,33,35,40} It is also documented that hemoglobin variants or the rate of hemoglobin glycation, which depends on erythrocyte properties, and red blood cell lifespan can affect A1C, although humans and rhesus macaques have comparable hematologic values, including red blood cell distribution width and disk diameter.^{13,24,33,36,41,42} There are reports, however, of structural differences in hemoglobin between different species of NHPs and between NHPs and humans.^43,44 In addition to the reasons listed above, some of the previous studies did not use point-of-care kits as we employed and measured A1C spectrophotometrically, by latex agglutination assay, chemiluminescence detection, or had samples frozen and shipped to commercial laboratories for analysis with the interval from collection to analysis up to 1 week, which differs from the combined immunoassay and general chemistry employed by the A1CNow⁺ kit used in this study.^{6,23,24,26,37,45} The A1CNow⁺ test kit also correlates closely with laboratory-derived standardized HPLC.^26,45,46

In our study we found that male rhesus macaques had higher A1C values than females, which aligns with differences noted in humans, although some results were dependent on age. This may be due to males having higher hemoglobin and iron levels than females.^{17,27,28,32,33}

In summary, we acknowledge that the results of our study identified A1C levels higher than other studies involving NHPs and higher levels than those described for humans. We sampled 210 rhesus macaques, more than any other published study that we found, and we used a commercially available point-of-care test kit that is certified by the National Glycohemoglobin Standardization Program and the International Federation of Clinical Chemistry and Laboratory Medicine. As discussed above, we believe that differences in species may contribute to the differences in A1C values, as variations in human A1C levels are found different ethnic or demographic groups. Another reason for the variation in A1C levels could be with respect to how the samples were analyzed. Only one study, McTighe et al,³⁸ used a point-of-care test kit, and all other studies had more extensive testing that normally requires an outside laboratory. A weakness of our study is that we did not compare the A1C results between the point-of-care HbA1c test kit, A1CNow⁺, and those from a commercial laboratory. This could have confirmed the findings from Johnston et al,³⁹ who found that the point-of-care kit had consistently higher values than those from commercial laboratories. However, we believe that our results are still valuable, as point-of-care A1C kits appear to be a much more convenient option for assessing A1C in NHPs.

Moving forward, we plan to investigate the A1C values of aged animals. The TNPRC has a National Institute of Aging colony consisting of 45 animals. Most of these animals are housed in the TNPRC breeding colony, which would make the corresponding fasted glucose measurements difficult to obtain. We would also like to include animals considered to be prediabetic by metabolic syndrome standards and true diabetic rhesus macaques to establish those standard values.