Hypoglycemia laboratory findings
|
Hypoglycemia Microchapters |
|
Diagnosis |
|---|
|
Treatment |
|
Case Studies |
|
Hypoglycemia laboratory findings On the Web |
|
American Roentgen Ray Society Images of Hypoglycemia laboratory findings |
|
Risk calculators and risk factors for Hypoglycemia laboratory findings |
Editor-In-Chief: C. Michael Gibson, M.S., M.D. [1]
Laboratory Findings
In the majority of children and adults with recurrent, unexplained hypoglycemia, the diagnosis may be determined by obtaining a sample of blood during hypoglycemia. If this critical sample is obtained at the time of hypoglycemia, before it is reversed, it can provide information that would otherwise require a several-thousand-dollar hospital admission and unpleasant starvation testing. Perhaps the most common inadequacy of emergency department care in cases of unexplained hypoglycemia is the failure to obtain at least a basic sample before giving glucose to reverse it.
- Glucose: needed to document actual hypoglycemia
- Insulin: any detectable amount is abnormal during hypoglycemia, but physician must know assay characteristics
- Cortisol: should be high during hypoglycemia if pituitary and adrenals are functioning normally
- Growth hormone: should rise after hypoglycemia if pituitary is functioning normally
- Electrolytes and total carbon dioxide: electrolyte abnormalities may suggest renal or adrenal disease; mild acidosis is normal with starvation hypoglycemia; usually no acidosis with hyperinsulinism
- Liver enzymes: elevation suggests liver disease
- Ketones: should be high during fasting and hypoglycemia; low levels suggest hyperinsulinism or fatty acid oxidation disorder
- Beta-hydroxybutyrate: should be high during fasting and hypoglycemia; low levels suggest hyperinsulinism or fatty acid oxidation disorder
- Free fatty acids: should be high during fasting and hypoglycemia; low levels suggest hyperinsulinism; high with low ketones suggests fatty acid oxidation disorder
- Lactic acid: high levels suggest sepsis or an inborn error of gluconeogenesis such as glycogen storage disease
- Ammonia: if elevated suggests hyperinsulinism due to glutamate dehydrogenase deficiency, Reye syndrome, or certain types of liver failure
- C-peptide: should be undetectable; if elevated suggests hyperinsulinism; low c-peptide with high insulin suggests exogenous (injected) insulin
- Proinsulin: detectable levels suggest hyperinsulinism; levels disproportionate to a detectabe insulin level suggest insulinoma
- Ethanol: suggests alcohol intoxication
- Toxicology screen: can detect many drugs causing hypoglycemia, especially for sulfonylureas
- Insulin antibodies: if positive suggests repeated insulin injection or antibody-mediated hypoglycemia
- Urine organic acids: elevated in various characteristic patterns in several types of organic aciduria
- Carnitine, free and total: low in certain disorders of fatty acid metabolism and certain types of drug toxicity and pancreatic disease
- Thyroxine and TSH: low T4 without elevated TSH suggests hypopituitarism or malnutrition
- Acylglycine: elevation suggests a disorder of fatty acid oxidation
- Epinephrine: should be elevated during hypoglycemia
- Glucagon: should be elevated during hypoglycemia
- IGF-1: low levels suggest hypopituitarism or chronic malnutrition
- IGF-2: low levels suggest hypopituitarism; high levels suggest non-pancreatic tumor hypoglycemia
- ACTH: should be elevated during hypoglycemia; unusually high ACTH with low cortisol suggests Addison's disease*
- Alanine or other plasma amino acids: abnormal patterns may suggest certain inborn errors of amino acid metabolism or gluconeogenesis
Defining Hypoglycemia
The precise level of glucose considered low enough to define hypoglycemia is dependent on (1) the measurement method, (2) the age of the person, (3) presence or absence of effects, and (4) the purpose of the definition. While there is no disagreement as to the normal range of blood sugar, debate continues as to what degree of hypoglycemia warrants medical evaluation or treatment, or can cause harm.[1][2][3] This article expresses glucose in milligrams per deciliter (mg/dL or mg/100 mL) as is customary in the United States, while millimoles per liter (mmol/L or mM) are the SI (International System) units used in most of the rest of the world. Glucose concentrations expressed as mg/dL can be converted to mmol/L by dividing by 18. For example, a glucose concentration of 90 mg/dL is 5 mmol/L or 5 mM.
Measurement Method
Blood glucose levels discussed in this article are venous plasma or serum levels measured by standard, automated glucose oxidase methods used in medical laboratories. For clinical purposes, plasma and serum levels are similar enough to be interchangeable. Arterial plasma or serum levels are slightly higher than venous levels, and capillary levels are typically in between.[4] This difference between arterial and venous levels is small in the fasting state but is amplified and can be greater than 10% in the postprandial state.[5] On the other hand, whole blood glucose levels (e.g., by fingerprick meters) are about 10%-15% lower than venous plasma levels.[4] Furthermore, available finger stick glucose meters are only warranted to be accurate to within 15% of a simultaneous laboratory value under optimal conditions, and home use in the investigation of hypoglycemia is fraught with misleading low numbers.[6][7] In other words, a meter glucose reading of 39 mg/dL could be properly obtained from a person whose laboratory serum glucose was 53 mg/dL; even wider variations can occur with "real world" home use.
Two other factors significantly affect glucose measurement: hematocrit and delay after phlebotomy. The disparity between venous and whole blood concentrations is greater when the hematocrit is high,[8] as in newborn infants, or adults with polycythemia. High neonatal hematocrits are particularly likely to confound glucose measurement by meter. Second, unless the specimen is drawn into a fluoride tube or processed immediately to separate the serum or plasma from the cells, the measurable glucose will be gradually lowered by in vitro metabolism of the glucose at a rate of approximately 7 mg/dL/hr, or even more in the presence of leukocytosis.[9][10][11]
Age Differences
Surveys of healthy children and adults show that plasma glucoses below 60 mg/dL (3.3 mM) or above 100 mg/dL (5.6 mM) are found in less than 5% of samples after an overnight fast.[12] In infants and young children up to 10% have been found to be below 60 mg/dL after an overnight fast. As the duration of fasting is extended, plasma glucose levels can fall further, even in healthy people. In other words, many healthy people can occasionally have glucose levels in the hypoglycemic range without symptoms or disease.
The normal range of newborn blood sugars continues to be debated. Surveys and experience have revealed blood sugars often below 40 mg/dL (2.2 mM), rarely below 30 mg/dL (1.7 mM) in apparently healthy full-term infants on the first day after birth. It has been proposed that newborn brains are able to use alternate fuels when glucose levels are low more readily than adults. Experts continue to debate the significance and risk of such levels, though the trend has been to recommend maintenance of glucose levels above 60-70 mg/dL after the first day after birth. In ill, undersized, or premature newborns, low blood sugars are even more common, but there is a consensus that sugars should be maintained at least above 50 mg/dL (2.8 mM) in such circumstances. Some experts advocate 70 mg/dL as a therapeutic target, especially in circumstances such as hyperinsulinism where alternate fuels may be less available.
Presence or Absence of Effects
Research in healthy adults shows that mental efficiency declines slightly but measurably as blood glucose falls below 65 mg/dL (3.6 mM) in many people. Hormonal defense mechanisms (adrenaline and glucagon) are activated as it drops below a threshold level (about 55 mg/dL for most people), producing the typical symptoms of shakiness and dysphoria. On the other hand, obvious impairment does not often occur until the glucose falls below 40 mg/dL, and up to 10% of the population may occasionally have glucose levels below 65 in the morning without apparent effects. Brain effects of hypoglycemia, termed neuroglycopenia, determine whether a given low glucose is a "problem" for that person, and hence some people tend to use the term hypoglycemia only when a moderately low glucose is accompanied by symptoms.
Even this criterion is complicated by the facts that A) hypoglycemic symptoms are vague and can be produced by other conditions; B) people with persistently or recurrently low glucose levels can lose their threshold symptoms so that severe neuroglycopenic impairment can occur without much warning; and C) many of our measurement methods (especially glucose meters) are imprecise at low levels.
Diabetic hypoglycemia represents a special case with respect to the relationship of measured glucose and hypoglycemic symptoms for several reasons. Although home glucose meter readings are sometimes misleading, the probability that a low reading accompanied by symptoms represents real hypoglycemia is higher in a person who takes insulin. Second, the hypoglycemia has a greater chance of progressing to more serious impairment if not treated, compared to most other forms of hypoglycemia that occur in adults. Third, because glucose levels are above normal most of the time in people with diabetes, hypoglycemic symptoms may occur at higher thresholds than in people who are normoglycemic most of the time. For all of these reasons, people with diabetes usually use higher meter glucose thresholds to determine hypoglycemia.
Purpose of Definition
For all of the reasons explained in the above paragraphs, deciding whether a blood glucose in the borderline range of 45-75 mg/dL (2.5-4.2 mM) represents clinically problematic hypoglycemia is not always simple. This leads people to use different "cutoff levels" of glucose in different contexts and for different purposes.
References
- ↑ Koh TH, Eyre JA, Aynsley-Green A (1988). "Neonatal hypoglycaemia--the controversy regarding definition". Arch. Dis. Child. 63 (11): 1386–8. PMID 3202648.
- ↑ Cornblath M, Schwartz R, Aynsley-Green A, Lloyd JK (1990). "Hypoglycemia in infancy: the need for a rational definition. A Ciba Foundation discussion meeting". Pediatrics. 85 (5): 834–7. PMID 2330247.
- ↑ Cornblath M, Hawdon JM, Williams AF, Aynsley-Green A, Ward-Platt MP, Schwartz R, Kalhan SC (2000). "Controversies regarding definition of neonatal hypoglycemia: suggested operational thresholds". Pediatrics. 105 (5): 1141–5. PMID 10790476.
- ↑ 4.0 4.1 Tustison WA, Bowen AJ, Crampton JH (1966). "Clinical interpretation of plasma glucose values". Diabetes. 15 (11): 775–7. PMID 5924610.
- ↑ [edited by] John Bernard Henry (1979). Clinical diagnosis and management by laboratory methods. Philadelphia: Saunders. ISBN 0-7216-4639-5.
- ↑ Clarke WL, Cox D, Gonder-Frederick LA, Carter W, Pohl SL (1987). "Evaluating clinical accuracy of systems for self-monitoring of blood glucose". Diabetes Care. 10 (5): 622–8. PMID 3677983.
- ↑ Gama R, Anderson NR, Marks V (2000). "'Glucose meter hypoglycaemia': often a non-disease". Ann. Clin. Biochem. 37 ( Pt 5): 731–2. PMID 11026531.
- ↑ [edited by] John Bernard Henry (1979). Clinical diagnosis and management by laboratory methods. Philadelphia: Saunders. ISBN 0-7216-4639-5.
- ↑ de Pasqua A, Mattock MB, Phillips R, Keen H (1984). "Errors in blood glucose determination". Lancet. 2 (8412): 1165. PMID 6150231.
- ↑ Horwitz DL (1989). "Factitious and artifactual hypoglycemia". Endocrinol. Metab. Clin. North Am. 18 (1): 203–10. PMID 2645127.
- ↑ [edited by] John Bernard Henry (1979). Clinical diagnosis and management by laboratory methods. Philadelphia: Saunders. ISBN 0-7216-4639-5.
- ↑ Samuel Meites, editor-in-chief; contributing editors, Gregory J. Buffone... [et al.] (1989). Pediatric clinical chemistry: reference (normal) values. Washington, D.C: AACC Press. ISBN 0-915274-47-7.