When collecting specimens for a GTT The most important information after the patients name is?

Early Screening for Overt Diabetes and Detection of Gestational Diabetes Mellitus

The frequency of diabetes complicating pregnancy has generally been estimated to be as high as 6% to 7%, with approximately 90% of these cases representing women with GDM.43 Using a pregnancy risk-assessment monitoring system between 2007 and 2010, DeSisto and colleagues43 confirmed that the prevalence of GDM is rising in the United States and may now be as high as 9.2%. The increasing frequency of GDM can be attributed to the worldwide obesity epidemic, as well the introduction of less stringent criteria for the diagnosis now being adapted by some institutions. An increased prevalence of GDM is found in women of ethnic groups that have high frequencies of type 2 diabetes. These include women of Hispanic, African, Native American, Asian, and Pacific Island ancestry. Women with GDM represent a group with significant risk for developing glucose intolerance later in life. O'Sullivan projected that 50% of women with GDM would become diabetic in follow-up at 22 to 28 years. Kjos and colleagues reported that 60% of Latina women will develop type 2 diabetes, and this level of risk may actually be manifest by 5 years after the GDM index pregnancy. The likelihood for subsequent diabetes increases when GDM is diagnosed in early pregnancy and when maternal fasting glucose levels are elevated. Presumably, some of these women with impaired β-cell function may represent cases of unidentified preexisting type 2 diabetes.

As noted earlier, GDM is a state restricted to pregnant women whose impaired glucose tolerance (IGT) is discovered during pregnancy. Because in most cases, patients with GDM have normal fasting glucose levels, some challenge of glucose tolerance must be undertaken. In the past, obstetricians relied on historic and clinical risk factors to select those patients most likely to develop GDM. This group included patients with a family history of diabetes and those whose past pregnancies were marked by an unexplained stillbirth or the delivery of a malformed or macrosomic infant. Obesity, hypertension, glycosuria, and maternal age older than 25 years were other indications for screening. Although multiple risk factors may increase the likelihood of GDM, more than half of all women who exhibit an abnormal GTT lack any of these risk factors.

The ACOG suggests that all pregnant women be screened for GDM, whether by the patient's medical history, clinical risk factors, or laboratory screening tests to determine blood glucose levels (Box 45.2). Screening is generally performed at 24 to 28 weeks’ gestation, when the “diabetogenic state” of pregnancy has been established. Boththe ADA and ACOG recommend early pregnancy screening for undiagnosed type 2 diabetes in women with a previous history of GDM, women who have delivered an infant weighing 4000 g or more, women with a history of impaired glucose metabolism or cardiovascular disease, those who are overweight or obese (BMI ≥25), women with a history of hypertension or polycystic ovary syndrome or a first-degree relative with diabetes, and members of high-risk race or ethnicity groups as noted previously. However, studies validating various approaches to early pregnancy screening for both type 2 diabetes and GDM are lacking. A simple approach to early screening for diabetes is to obtain an HbA1c level in women believed to be at greatest risk. A diagnosis of overt diabetes is made if the level is greater than or equal to 6.5%. In these women who may have had diabetes during embryogenesis, a comprehensive ultrasound should be performed later in pregnancy to detect fetal malformations. If the level is between 5.7% and 6.4%, suggesting IGT, a diagnostic oral GTT is performed. Fong and colleagues have reported that 25 percent of these women will subsequently develop GDM. For those women with normal results or those with an HbA1c level less than 5.7%, screening for GDM is then undertaken at 24 to 28 weeks. Recently, Hughes and colleagues44 suggested that an early HbA1c level greater than or equal to 5.9% is optimal for detecting overt diabetes in women at less than 20 weeks’ gestation and also identifies women at increased risk of adverse pregnancy outcomes.

2 Glucose Tolerance and the Insulin Response

The glucose tolerance test is the most important test of carbohydrate function and is of particular value in those cases of diabetes in which the fasting blood glucose level is only moderately elevated and the diagnosis is equivocal. The blood glucose curve in diabetes characteristically shows a decreased tolerance for glucose (Fig. 12), as evidenced by the long T½ or low k value, which reflects the inability of the animal to dispose of a test dose of glucose. The insulin response curve in Type I diabetes clearly demonstrates the inability of the pancreas to release insulin in response to the glucose load. The absence of an insulin response plays the primary role in the failure of the diabetic animal to utilize the added glucose. An equally important contributory factor is the overproduction of glucose by the diabetic liver in the presence of hyperglycemia. The test dose of glucose is in effect added to the burden of an already existing oversupply of glucose. Since the steady-state level at which the liver ceases to supply or remove glucose is elevated in diabetes, the liver continues to supply glucose, which contributes to the delayed return of the tolerance curve to its original level.

In Types II and III diabetes (see below), there is also glucose intolerance, but this occurs in the presence of a normal to elevated immunoreactive insulin (IRI) level. This would mean that the IRI in the plasma of these types is ineffective due to insulin resistance, binding to antibodies or other proteins, or abnormal structure or that a cross-reacting insulin-like molecule such as C-peptide is being measured. Thus, the glucose tolerance curve reflects an absolute (Type I) or relative (Types II, III) lack of insulin.

Gestational Diabetes

GDM is defined as glucose intolerance diagnosed during pregnancy that is not clearly preexisting diabetes. The underlying pathophysiology of GDM in most instances is similar to that observed for T2DM: an inability to maintain an adequate insulin response because of the significant decreases in insulin sensitivity with advancing gestation.

It is estimated that 5% to 9% of pregnant women will be diagnosed with GDM using the standard two-step procedure based on the 100-g, 3-hour OGTT.19 This is not surprising, because GDM is the precursor to T2DM, based on ethnicity-specific increases in insulin resistance and rising obesity in women of reproductive age. Clinical recognition of GDM is important because therapy can reduce pregnancy complications and potentially reduce long-term sequelae in the offspring.

The use of the 3-hour, 100-g OGTT for diagnosis of GDM was based on John O'Sullivan's work in the 1960s.20 O'Sullivan established criteria for the OGTT that served as a risk predictor for development of T2DM after the pregnancy; however, the limits for the 3-hour 100-g OGTT were defined to yield a group of women in the top 98th percentile of glucose response. Other problems with use of the 3-hour OGTT—including the propensity for patient emesis after such a large glucose load, which invalidated the test—led to the development of the two-step testing regimen used widely in the United States, which involves a 50-g “challenge” pretest followed by the 100-g OGTT only if the initial result is abnormal. However, the two-step diagnostic system inherently imposes delay in diagnosis of GDM and has a relatively high false-negative rate (10% to 20%).

In 2010, the International Association of Diabetes and Pregnancy Study Groups (IADPSG) recommended new criteria for the diagnosis of GDM.21 These criteria (Box 59.2 andTable 59.1) are based on the results of a large (25,000 participants) multinational observational trial known as the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study,22 which meticulously tracked adverse pregnancy and neonatal outcomes. The new IADPSG criteria were based primarily on the risks of excessive adverse perinatal outcomes (odds ratios [ORs] >1.75) for newborn birth weight, body fat, and cord C-peptide level, adjusted for confounding variables.

The new IADPSG criteria for GDM diagnosis have been adopted widely globally (by the World Health Organization) but notably with limitations in the United States by the ADA,13 and not by the American College of Obstetricians and Gynecologists (ACOG),23 largely because of the substantial increase in women diagnosed with GDM—up to 18% using the new criteria, compared to 5% to 10% with the two-step 3-hour OGTT system—without clear evidence of benefit.

5 Glucose Tolerance and the Insulin Response

The glucose tolerance test (GTT) is the most important test of carbohydrate function and is of particular value in those cases of diabetes in which the fasting blood glucose is only moderately elevated and the diagnosis is equivocal (Section VIII.C). The diabetic oral GGT curve is high and relatively flat, indicating a decreased tolerance for glucose (Fig. 3-12). The nature of the diabetic curve can be quantitated by using the intravenous GTT. The diabetic curve is characterized by a long T1/2 or low k-value, which reflects the inability of the animal to use the test dose of glucose. The insulin response curve in type I (absolute insulin deficiency) diabetes clearly demonstrates the inability of the pancreas to release insulin in response to the glucose load. It is in the absence of an insulin response, which is responsible for the failure of the diabetic to utilize the added glucose, that the prolonged hyperglycemia occurs. An important factor adding to the hyperglycemia is the overproduction of glucose by the liver. The test dose of glucose is in effect added to the already existing oversupply of glucose. Because the steady-state level at which the liver ceases to supply or remove glucose is elevated in diabetes, the liver continues to oversupply glucose, which contributes to the slow return of the tolerance curve to its original level.

In types II and III diabetes (see the following discussion), there is also glucose intolerance, but this occurs in the presence of a normal to elevated insulin. This would mean that the insulin in the plasma of these types is unusable or ineffective (i.e., relative deficiency) because of a number of factors including insufficient receptors, receptor blockage, abnormal receptor structure, or antibody binding, all of which lead to the glucose intolerance and the phenomenon of insulin resistance. Therefore, glucose intolerance is seen in all types of diabetes whether there is an absolute (type I) or relative (types II, III) deficiency of insulin. The insulin response must be evaluated in order to establish the type of diabetes.

Diagnosis

A diagnosis of diabetes has historically included an elevated fasting blood glucose level, any glucose value higher than 200 mg/dL (11 mmol/L) with symptoms of hyperglycemia, or an abnormal 2-hour oral glucose tolerance test (OGTT).9 American Diabetes Association (ADA) guidelines for the diagnosis of diabetes were modified in 2009 to include a hemoglobin A1c (HbA1c) value greater than 6.5%.10 Under certain settings (e.g., obesity, racial status other than Caucasian) and particularly among adults, the diagnosis of T1DM versus type 2 diabetes mellitus (T2DM) can prove quite challenging. At present, the best criterion for separating the two disorders resides in laboratory identification of any one of a number of islet cell autoantibodies (also known as T1DM-associated autoantibodies; i.e., anti-insulin autoantibodies [IAA], anti–glutamic acid decarboxylase [GADA], anti–insulinoma-associated antigen 2 [IA2A], or anti-zinc-transporter 8 [ZnT8A]). Literally hundreds of studies over the past three decades have suggested that the presence of these autoantibodies provides high sensitivity for diagnosing persons with T1DM.11 Indeed, more than 90% of Caucasian children presenting with diabetes express at least one of these four T1DM-associated autoantibodies.12 In terms of disease specificity, T1DM-associated autoantibodies are typically positive in less than 1% to 2% of unaffected (i.e., non-T1DM) individuals, further validating their diagnostic utility. However, among African-American and Latino children and adolescents in the United States diagnosed with diabetes, almost one half lack any T1DM-associated autoantibody. Many patients from these ethnic minorities in the United States present clinically as if they have early-onset T2DM (e.g., mild ketosis, slow symptomatic onset), some have attendant risk factors such as obesity, and many lack human leukocyte antigen (HLA) alleles associated with T1DM. Increasingly diverse genetic admixtures, due to geographic migration and social changes (e.g., multiracial offspring), further contribute to diagnostic complexity.

In childhood and adolescence, two peaks of T1DM presentation occur: a smaller peak between 5 and 7 years of age and a larger peak at or near puberty.13 Although most autoimmune disorders disproportionately target females, T1DM affects males slightly more than females. The incidence of T1DM varies with seasonal changes and birth month. Incidence of T1DM diagnosis is higher in autumn and winter, whereas being born in the spring is associated with an increased likelihood for T1DM.14 Interestingly, the development of T1DM-associated autoimmunity (i.e., the formation of islet autoantibodies) in the months to years prior to the onset of symptomatic T1DM also shows a degree of seasonal synchronization.

The measurement and presence of T1DM-associated autoantibodies have also fueled much debate regarding the percentage of T1DM cases that are errantly misclassified as T2DM. Indeed, it is conceivable that 5% to 15% of adults diagnosed with T2DM may have T1DM, given the frequency of T1DM-associated autoantibodies in populations diagnosed with T2DM.15 This is, in effect, a problem in health care provider recognition of the potential for T1DM disease in adult populations combined with a lack of widespread screening for such autoantibodies in settings in which screening would seem warranted. If this assertion is correct, given the vastly greater number of persons diagnosed with T2DM (relative to that of T1DM), the number of actual T1DM cases in a given population may be massively underestimated. This is therapeutically unfortunate because an accurate diagnosis of T1DM is vital; persons misdiagnosed as having T2DM when they indeed have T1DM experience higher HbA1c, have risk of diabetic ketoacidosis (DKA) due to use of noninsulin therapies, and have accelerated progression toward micro- and macrovascular complications..

Laboratory test for diabetes mellitus

A glucose tolerance test is designed to assess the patient's metabolism when given oral carbohydrate under controlled conditions. The findings of the laboratory examination provide a graph of blood sugar levels at regular intervals, combined with this are the result of urine tests at the same times to show the points at which glucose filters through the kidney and produces glycosuria (see illustration 4:3).

Preparation of the patient includes an explanation of the procedure, its purpose and the method to be used. The nurse should aim at getting not just a passive obedience from the patient, but his active cooperation. In this instance it should be explained that the test is to confirm the diagnosis made by the patient's own doctor; that it will be necessary to take samples of the patient's blood and urine at regular intervals, and that if the test is to be a success he must not eat or drink anything unless told to do so.

The patient must take nothing by mouth after 9 pm the previous evening allowing the test to commence at 9 am the following morning, thus minimising the time that the patient is deprived of food. At 9 am a specimen of blood is taken to determine the patient's ‘fasting’ sugar level ie the level of a person who has not eaten for at least four to six hours. In normal subjects the fasting blood sugar is near the minimum level following a fast (60 to 120mg/100ml). The diabetic patient will not have such a low level and a degree of glycosuria is possible. Also at this time the patient is required to empty his bladder, a specimen being retained.

The patient is given 50gm of glucose by mouth. It is usual to mix it with some flavouring in 100ml of water, the result being more palatable. The patient should be at rest from this time and no smoking is allowed. From the time the glucose is given blood specimens are taken every half hour afterwards until four specimens are collected. The patient is required to pass a specimen of urine half hourly during the two hours. The blood specimens need to be collected in bottles containing a preservative, the amount to be collected is marked on the label. Care is necessary to ensure that each bottle is marked with the patient's name and the time collected. When specimens are collected they should be marked immediately to avoid mistakes. Careless marking or timing can result in a faulty graph and cause repetition of the procedure with added stress to the patient. All specimens are dispatched to the laboratory as soon as possible while the samples are fresh.

The result of the test is received in the form of a graph showing an estimation of blood sugar levels coupled with urinalysis. The curve of a normal subject starts at a lower level that the diabetic patient and will rise quickly during the first half hour, insulin then controls the blood sugar level by withdrawing any excess. Glycosuria will be negative in each specimen.

The curve of the diabetic patient starts at a much higher point in the scale and it quickly rises during the first half hour and then continues to rise at a slower rate. The fall of blood sugar does not reach the fasting reading until several hours have passed. Glycosuria may be positive from the first sample or during the first hour. Minor degrees of diabetes may be impossible to diagnose until the test for glucose tolerance is performed. Once confirmed treatment should begin as soon as possible.

Glucose tolerance test

The 75 g glucose tolerance test (GTT) remains the gold standard test for the diagnosis of acromegaly. After glucose, plasma GH in normal individuals is suppressed to <0.4 μg/L. Patients with acromegaly fail to show complete suppression and may show a paradoxical rise of plasma GH. Failure of GH to fall to <0.4 μg/L is highly suggestive of acromegaly, even if substantial suppression of GH has occurred, and an elevated IGF-1 will normally confirm the diagnosis.

Failure of normal glucose suppression of GH may occur in the absence of acromegaly in liver disease, poorly controlled diabetes, chronic kidney disease, malnutrition (including anorexia nervosa) and primary growth hormone insensitivity (Laron syndrome). Diabetes can be a particular diagnostic challenge, since a GTT is inappropriate in established disease. Insulin-like growth factor 1 concentrations, however, are typically normal in uncomplicated diabetes, and, if elevated, the authors usually recommend a four-point GH day curve to confirm circulating GH concentrations consistently above the ‘safe’ range.

2 Technique

Performance of the CutGTT has been described in detail (Fusaro and Johnson, 1965) and will be outlined only briefly here. The test is an extension of the IVGTT, with the additional feature that skin glucose concentration is also monitored. Triplicate blood and skin specimens were obtained from a fasting subject, and 35 gm of glucose was administered intravenously. At appropriate intervals blood specimens were collected and lysed in 30 vol of water. Likewise, skin specimens were obtained from the back with a motor-driven, 3-mm rotary punch, quickly weighed, and immersed in 1% aqueous ZnSO4. After 1.5 hours, an equimolar quantity of 0.06 N Ba(OH)2 was added to each specimen, and the mixtures were centrifuged. The four aliquots obtained from each clear supernatant fluid were employed in two separate glucose assays (in duplicate), with glucose oxidase reagent. Blood samples were depro-teinized with 0.3 N Ba(OH)2 and 5% ZnSO4, and aliquots were assayed with the microvolumetric procedure employed for skin. Excess blood and skin glucose values were calculated by subtracting average fasting glucose from the postinjection levels. The plots of logarithm excess glucose versus postinjection time were fitted with straight lines by the method of least squares; and from the slopes of the corresponding lines a disappearance rate constant (percentage per minute) for blood glucose and for skin glucose was calculated.

Interpretation

WHO criteria for diabetes and impaired glucose tolerance

The 75-gram oral glucose tolerance test.

In the absence of diabetic symptoms at least two abnormal values are necessary to establish a diagnosis of diabetes mellitus.

Diagnosis

Metabolic syndrome can be diagnosed using the criteria described in Table 32.1. Outside a research laboratory, standard of practice for the diagnosis of the metabolic syndrome is to conduct a 2-hour glucose and insulin tolerance test (GITT), which can easily be ordered through any outpatient laboratory. The protocol is as follows: (1) 2 days of carbohydrate loading, (2) blood sampling for fasting glucose and insulin measurements, and (3) consumption of a 75-g glucose drink. Thereafter, blood specimens for glucose and insulin measurements are obtained (but not always necessarily) at half-hour intervals for the first hour, followed by a final specimen 2 hours later. In the majority of patients, fasting and 2-hour measurements are sufficient. Essentially, this protocol is the standard glucose tolerance test (GTT) with concomitant insulin testing. A caveat of this approach is that clinicians should ensure the laboratory used is familiar with diagnostic procedures involving insulin, which is a very unstable hormone in vitro. False-negative diagnoses are possible if only a fasting insulin specimen is obtained. Regular drawing of blood specimens for the measurement of insulin is required to maintain test functionality. Clinician are required to understand the effect of insulin on managing blood glucose levels following glucose consumption. Baseline fasting insulin levels should normally be less than 15 microunits/mL and less than 30 microunits/mL at 2 hours following consumption of a 75-g glucose load.

Although the 2-hour GITT is considered the most accurate and functional test, other methods for the diagnosis of the metabolic syndrome include the following:

Triglyceride-to-HDL cholesterol ratio (TG:HDL-C)—a healthy ratio is less than 2.

Glycosylated hemoglobin (HbA1c): values for patients with IR are between 5.7 and 6.4 as per the 2010 American Diabetes Association guidelines.63

Fasting insulin: When assessed in isolation, normal values should be less than 15 microunits/mL (140 pmol/L); however, a normal fasting insulin result does not rule out IR. Reference ranges are laboratory specific, so clinicians must check with the clinical laboratories for specific values. In addition, what is “normal” and what is “healthy” can be vastly different.

Other markers of importance include elevated hs-CRP, uric acid, small dense low-density cholesterol (sd LDL-C), and inflammatory markers such as IL-6 and IL-8, TNF-alpha, PAF-1, and adiponectin. Because NAFLD is hypothesized to represent the hepatic manifestation of IR, the measurement of gamma-glutamyl transpeptidase (GGT) levels should also be considered as this transaminase enzyme is the most sensitive in detecting liver toxicity.

For the majority of clinicians, the 2-hour GITT is the most valuable in terms of diagnosis and patient education, particularly for normal-weight individuals suspected to have metabolic syndrome and women with polycystic ovarian syndrome.