Early detection of diabetes. Screening for diabetes
Recommendations For all women at the 24-28th week of pregnancy, it is recommended to perform an oral glucose tolerance test to detect diabetes in pregnant women . Routine screening in asymptomatic non-pregnant adults with blood glucose or urinalysis is not recommended, but screening is indicated for some groups with an increased risk factor (see Clinical intervention).
About 11 million people in the United States have diabetes . Diabetes can cause life-threatening complications and is a significant risk factor for other deadly diseases, such as coronary artery disease, congestive heart disease, and cerebral artery damage. Diabetes is the seventh leading cause of death in the United States — more than 130,000 deaths a year, also causing death from other diseases.
Diabetes is the main cause of neuropathy, which develops in at least 50% of patients up to 25 years after the first diagnosis of diabetes. Diabetic damage to peripheral vessels entails more than 50 thousand amputations per year6. Microvascular lesions from diabetes cause kidney damage and blindness. Diabetic nephropathy – a complication that occurs in about 10% of cases of diabetes, gives about a quarter of new dialysis patients.
Diabetes is a leading cause of blindness in adults, and as a result of this disease, about 5800 people lose their vision every year. Newborns born to mothers with diabetes are at increased risk for preterm birth, perinatal mortality, overweight, congenital malformations and metabolic abnormalities. Direct and indirect diabetes losses in the United States are at least $ 14 billion a year.
About 90% of all cases of diabetes are type 2, i.e., non-insulin -dependent type of diabetes mellitus (NIDDM). This form of diabetes usually develops in adults and more often appears after 40 years. Diabetes has been reported in 2 million older Americans. NIDDM is especially often found in blacks, Hispanics, Indians. About 1 million black Americans have diabetes. Another significant risk of NIDDM is a family history of diabetes, obesity, and diabetes during labor. Type 1 diabetes mellitus or IDDM accounts for approximately 10% of all cases of diabetes, it is especially characteristic exacerbates in childhood or adolescence.
Pregnant diabetes – poor glucose tolerance during pregnancy in non-diabetic women – occurs in 3% of all pregnancies. This condition is a risk factor for macrosomalia (an increase in the size of the fetus), which may be associated with other complications of the mother and the newborn. Although macrosomalia in itself is not a pathological factor, it nevertheless increases the risk of birth trauma, disorders of the cranium and clavicle, displacement of the shoulder joint, and damage to the peripheral nerves. As mentioned, the presence of diabetes during childbirth means an increased risk of NIDDM in the mother, and can also be the forerunner of prolonged poor glucose tolerance.
The effectiveness of diabetes screening tests.
Although there are a number of different methods for checking for diabetes (for example, hemoglobin A1C), measuring hemoglobin in the blood is the main way to detect diabetes in asymptomatic individuals. Glucose can be measured randomly, on an empty stomach, after eating, or at predetermined intervals after a specific oral dose of glucose has been administered (glucose tolerance test (TGP). These tests are used to detect impaired glucose tolerance, a condition typical of diabetes, but which may occur and before diabetes. In order to state diabetes, and not just impaired glucose tolerance, we take as a criterion glucose in plasma of fasted blood of 140 mg / dl (7.8 mmol / l) or more, an increased level of hl plasma glucose after tolerance test of 75 g of orally administered glucose (200 mg / dl (11.1 mmol / L) or higher for 0–2 hours and 2 hours) or the presence of classic symptoms such as polyuria, polydipsia and ketonuria.
Since 1960, higher doses, 100 g and various threshold criteria have been used to detect the presence of diabetes in pregnant women. It is believed that in the absence of diabetes, impaired glucose tolerance should be considered if the plasma glucose level is from 140 mg / dl (7.8 mmol / l) to 200 mg / dl (11.1 mmol / l) 2 hours after giving 75 g of glucose and if for up to 120 minutes the plasma glucose level exceeds 200 mg / dl (11.1 mmol / l).
The need for such complex criteria is partly related to the difficulties of using a single determination of glucose as a basis for diabetes screening . There is no specific glucose level that could be used as the borderline separating individuals with impaired glucose tolerance, diabetes and normal. The ranges of blood glucose concentration in these three groups overlap. Moreover, even in one person, at different times and depending on the food taken, the level of glucose in the blood can vary greatly. Thus, if we take a low threshold level as a criterion for determining hyperglycemia, we will result in high sensitivity and poor specificity for detecting poor glucose tolerance and diabetes.
On the other hand, an undeniable sign of poor glucose tolerance is a high blood glucose level above 200 mg / dl (11.1 mmol / L), but if you take such a high threshold level for screening, then many cases of the disease will go unnoticed.
Each glucose screening test has its own advantages and disadvantages. Determination of glucose in a fasting blood sample is less practical for routine screening than random sampling, because in this case a person should not eat for 8-10 hours, but this test is more accurate. Nevertheless, its sensitivity as with the curing method is limited; one study showed that only 25% of people with undiagnosed diabetes had fasting blood glucose levels above 140 mg / dL (7.8 mmol / L). Checking glucose levels after meals (levels above 200 mg / dl (11.1 mmol / L) one and a half to two hours after eating may be more convenient for people and more sensitive to detect impaired glucose tolerance, but this method is not ideal in as a screening method due to time constraints.
The greatest accuracy is given by a test in which 75 g of glucose (TPG) is taken orally, but this test is not suitable for screening , it is inconvenient and expensive, since it is necessary to take a large amount of glucose and perform venous puncture several times over several hours. TPG is often used not as a screening test, but as a confirmatory method when diabetes is suspected.
TPG is used as a screening for diabetes of pregnant women with reduced time (1 hour) and with lower doses (50 g). The level of 140 mg / dl (7.8 mmol / L) or higher one hour after taking 50 g of glucose has indicators of 83% sensitivity and 87% specificity compared with 100 g glucose. If we take the 3% figure as an indicator of the frequency of diabetes in pregnant women, then the routine use of a test of 50 g in pregnant women will give five false-positive cases for every true case of diabetes in pregnant women. The test also reveals limited reproducibility – up to 75% of patients with positive TPG in the subsequent tests gave negative results. A large number of false-positive results cannot be discounted, as patients who are diagnosed with diabetes become suspicious, and then are forced to undergo additional diagnostic checks.
Restrictions in food and excessive monitoring of the fetus and childbirth – these are also side effects of a mistake in the diabetes test , which was not found in subsequent checks. True, any statistics that would document these facts are practically not available.
Using a urine test to detect glucosuria is considered an unsuitable screening test for diabetes because urine glucose levels fluctuate and glucosuria can also occur with normal blood glucose levels in individuals with a low renal threshold for glucose. Determination of glucose in urine has a sensitivity of less than 30%. In addition, the analysis is not accurate enough if the collection and storage of urine were not carried out correctly. Even in people with confirmed diabetes, a urine test is replaced by self-monitoring of blood glucose, considering it a more effective method for daily glycemic control.