Diabetes Mellitus and Disorders of Glucose Homeostasis

Chapter 126


Diabetes Mellitus and Disorders of Glucose Homeostasis




Diabetes Mellitus



Perspective


Diabetes mellitus is the most common endocrine disease. It comprises a heterogeneous group of hyperglycemic disorders characterized by high serum glucose concentration and disturbances of carbohydrate and lipid metabolism. Acute complications include hypoglycemia, diabetic ketoacidosis (DKA), and hyperglycemic hyperosmolar state (HHS). Long-term complications affect multiple organ systems through involvement of the microvasculature and include retinopathy, nephropathy, neuropathy, and angiopathy. As a result, complications such as coronary and cerebral vascular disease, blindness, chronic kidney disease, complicated infections, and amputations are present in a much higher incidence in diabetics than in nondiabetics. Diabetes is frequently ranked as one of the five major chronic diseases that account for a significant proportion of our health care spending. Several trials have shown to varying degrees that tight glucose control can reduce risk of death and several microvascular complications. Patients with diabetes mellitus incur emergency department (ED) costs three times higher than those of nondiabetic patients and are admitted to the hospital four times more often.



Principles of Disease



Normal Physiology


Maintenance of the plasma glucose concentration is critical to survival because plasma glucose is the predominant metabolic fuel used by the central nervous system (CNS). The CNS cannot synthesize glucose, store more than a few minutes’ supply, or concentrate glucose from the circulation. Brief hypoglycemia can cause profound CNS dysfunction, and prolonged severe hypoglycemia may cause cellular death. Glucose regulatory systems have evolved to prevent or to correct hypoglycemia.


The plasma glucose concentration is normally maintained within a relatively narrow range, between 60 and 150 mg/dL, despite wide variations in glucose levels after meals and exercise. Glucose is derived from three sources: intestinal absorption from the diet; glycogenolysis, the breakdown of glycogen; and gluconeogenesis, the formation of glucose from precursors, including lactate, pyruvate, amino acids, and glycerol. After glucose ingestion, the plasma glucose concentration increases as a result of glucose absorption. Endogenous glucose production is suppressed. Plasma glucose then rapidly declines to a level below the baseline.



Insulin.: Insulin receptors on the beta cells of the pancreas sense elevations in blood glucose concentration and trigger insulin release into the blood. For incompletely understood reasons, glucose taken by mouth evokes more insulin release than parenteral glucose does. Certain amino acids induce insulin release and even cause hypoglycemia in some patients. Sulfonylurea oral hypoglycemic agents work, in part, by stimulating the release of insulin from the pancreas.


The number of receptor sites helps determine the sensitivity of the particular tissue to circulating insulin. The number and sensitivity of receptor sites are also the primary factors in the long-term efficacy of the sulfonylurea oral hypoglycemic agents. Receptor sites are increased in glucocorticoid deficiency and may be relatively decreased in obese patients.


Under normal circumstances, insulin is rapidly degraded through the liver and kidneys. The half-life of insulin is 3 to 10 minutes in the circulation. Whereas insulin is the major anabolic hormone pertinent to the diabetic disorder, glucagon plays the role of the major catabolic hormone in disordered glucose homeostasis.


Although most tissues have the enzyme systems required to synthesize and to hydrolyze glycogen, only the liver and kidneys contain glucose-6-phosphatase, the enzyme necessary for the release of glucose into the circulation. The liver is essentially the sole source of endogenous glucose production. Renal gluconeogenesis and glucose release contribute substantially to the systemic glucose pool only during prolonged starvation.


The hepatocyte does not require insulin for glucose to cross the cell membrane. However, insulin augments both the hepatic glucose uptake and storage needed for the process of energy generation and glycogen and fat synthesis. Insulin inhibits hepatic gluconeogenesis and glycogenolysis.


Muscle can store and use glucose, primarily through glycolysis to pyruvate, which is reduced to lactate or transaminated to form alanine. Lactate released from muscle is transported to the liver, where it serves as a gluconeogenic precursor. Alanine may also flow from muscle to liver. During fasting, muscle can reduce its glucose uptake, oxidize fatty acids for its energy needs, and, through proteolysis, mobilize amino acids for transport to the liver as gluconeogenic precursors. Adipose tissue can also use glucose for fatty acid synthesis for oxidation to form triglycerides. During fasting, adipocytes can also decrease their glucose use and satisfy energy needs through the beta oxidation of fatty acids. Other tissues do not have the capacity to decrease glucose use on fasting and therefore produce lactate at relatively fixed rates.


Glucose transport across the fat cell membrane also requires insulin. A large percentage of the adipocyte glucose is metabolized to form α-glycerophosphate, required for the esterification of fatty acids to form triglycerides. Although most insulin-mediated fatty acid synthesis occurs in the liver, a very small percentage occurs in fat cells, with use of the acetyl coenzyme A generated by glucose metabolism. Very low levels of insulin are required to inhibit intracellular lipolysis while stimulating the extracellular lipolysis required for circulating lipids to enter the fat cell.



Glucose Regulatory Mechanisms.: Maintenance of the normal plasma glucose concentration requires precise matching of glucose use and endogenous glucose production or dietary glucose delivery. The regulatory mechanisms that maintain systemic glucose balance involve hormonal, neurohumoral, and autoregulatory factors. Glucose regulatory hormones include insulin, glucagon, epinephrine, cortisol, and growth hormone. Insulin is the main glucose-lowering hormone. Insulin suppresses endogenous glucose production and stimulates glucose use. Insulin is secreted from the beta cells of the pancreatic islets into the hepatic portal circulation and has important actions on the liver and the peripheral tissues. Insulin stimulates glucose uptake, storage, and use by other insulin-sensitive tissues, such as fat and muscle.


Counter-regulatory hormones include glucagon, epinephrine, norepinephrine, growth hormone, and cortisol. When glucose is not transported into the cells because of either a lack of food intake or a lack of insulin, the body perceives a “fasting state” and releases glucagon, attempting to provide the glucose necessary for brain function. In contrast to the fed state, in the fasting state the body metabolizes protein and fat. Glucagon is secreted from the alpha cells of the pancreatic islets into the hepatic portal circulation. Glucagon lowers hepatic levels of fructose 2,6-bisphosphate, resulting in decreased glycolysis and increased gluconeogenesis, an effect that may be enhanced by ketosis. Glucagon increases the activity of adenylate cyclase in the liver, thereby increasing glycogen breakdown to glucose and further increasing hepatic gluconeogenesis. Glucagon acts to increase ketone production in the liver. Thus, whereas insulin is an anabolic agent that reduces blood glucose concentration, glucagon is a catabolic agent that increases blood glucose concentration. Glucagon is released in response to hypoglycemia as well as to stress, trauma, infection, exercise, and starvation. It increases hepatic glucose production within minutes, although transiently.


Epinephrine both stimulates hepatic glucose production and limits glucose use through both direct and indirect actions mediated by both alpha-adrenergic and beta-adrenergic mechanisms. Epinephrine also acts directly to increase hepatic glycogenolysis and gluconeogenesis. It acts within minutes and produces a transient increase in glucose production but continues to support glucose production at approximately basal levels thereafter. Norepinephrine exerts hyperglycemic actions by mechanisms similar to those of epinephrine, except that norepinephrine is released from axon terminals of sympathetic postganglionic neurons.


Growth hormone initially has a plasma glucose–lowering effect. Its hypoglycemic effect does not appear for several hours. Thus, growth hormone release is not critical for rapid glucose counter-regulation; this is also true for cortisol. In the long term, both growth hormone and cortisol may also increase glucose production.



Types of Diabetes


The American Diabetes Association (ADA) defines four major types of diabetes mellitus: type 1 diabetes mellitus, type 2 diabetes mellitus, gestational diabetes, and diabetes due to secondary disease processes or drugs. The 1997 National Diabetes Data Group report discontinued the use of the terms insulin-dependent diabetes mellitus and non–insulin-dependent diabetes mellitus because they are confusing and clinically inaccurate. In addition, use of Arabic numerals (1 and 2) instead of Roman numerals is the standard. The most recent update to the standards of care for diabetes was published in January 2011.1 The diagnostic criteria for diagnosis of diabetes were changed in 2010 from the previous standards of elevated fasting glucose concentration and abnormal result of the 2-hour oral glucose tolerance test (OGTT) to use of the hemoglobin A1c (HbA1c) value as the preferred confirmatory test.1 An HbA1c value above 6.5% is now considered diagnostic of diabetes. However, the fasting plasma glucose concentration and 2-hour OGTT are still considered valid, as is the presence of a random glucose measurement of more than 200 mg/dL in a nonfasting patient. In addition, the use of fasting plasma glucose concentration may help identify patients at risk for diabetes (if their glucose concentration is elevated but not crossing the threshold for diagnosis of diabetes).







Impaired Glucose Tolerance.: Impaired glucose tolerance (IGT) and its analogue, impaired fasting glucose (IFG), are considered to identify individuals at high risk for development of diabetes. This group is composed of persons whose plasma glucose levels are between normal and diabetic and who are at increased risk for the development of diabetes and cardiovascular disease. The pathogenesis is thought to be related to insulin resistance. Presentations of IGT/IFG include nonketotic hyperglycemia, insulin resistance, hyperinsulinism, and often obesity.


IGT/IFG differs from the other classes in that it is not associated with the same degree of complications of diabetes mellitus. Many of these patients even spontaneously have normal glucose tolerance. One should not be complacent about the patient with IGT because the decompensation of this group into the category of diabetes mellitus is 1 to 5% per year.1



Epidemiology


The prevalence of diabetes is difficult to determine because many standards have been used. Regardless, the most recent data estimate that 8.3% of Americans of all ages and 11.3% of all adults older than 20 years have diabetes.1 Approximately 215,000 Americans younger than 20 years have diabetes. Of these, 5 to 10% have type 1, and 90 to 95% have type 2; other types account for 1 to 5% of cases.


The peak age at onset of type 1 diabetes is 10 to 14 years. Approximately 1 of every 600 schoolchildren has this disease. In the United States the prevalence of type 1 is approximately 0.26% by the age of 20 years, and the lifetime prevalence approaches 0.4%. The annual incidence among persons from birth to 16 years of age in the United States is 12 to 14 per 1 million population. The incidence is age dependent, increasing from near-absence during infancy to a peak occurrence at puberty and another small peak at midlife.1


The morbidity in diabetes is related primarily to its vascular complications. A mortality rate of 36.8% has been related to cardiovascular causes, 17.5% to cerebrovascular causes, 15.5% to diabetic comas, and 12.5% to renal failure.



Pathophysiology and Etiology


Type 1 diabetes results from a chronic autoimmune process that usually exists in a preclinical state for years. The classic manifestations of type 1—hyperglycemia and ketosis—occur late in the course of the disease, an overt sign of beta cell destruction. The most striking feature of long-standing type 1 diabetes is the nearly total lack of insulin-secreting beta cells and insulin, with the preservation of glucagon-secreting alpha cells, somatostatin-secreting delta cells, and pancreatic polypeptide-secreting cells.


Although the exact cause of diabetes remains unclear, research has provided many clues. Studies of the pathogenesis of diabetes mellitus have demonstrated that the cause of the disordered glucose homeostasis varies from individual to individual. This cause may determine the presentation in each patient. Individual patients are currently not studied for the source of their disease except on an experimental basis. The goals of the work in progress, however, are to identify who is susceptible to the development of diabetes and to prevent diabetic emergencies and sequelae or to prevent expression of the disease.


A genetic basis for diabetes is suggested by the association of type 1 disease with certain HLA markers and by the findings of numerous twin and family studies. Families who move from areas with a low frequency of type 1 diabetes to areas with a high frequency have an incidence of disease similar to that in the areas where they reside; this suggests an environmental basis for diabetes. An autoimmune cause has been clearly demonstrated in many type 1 diabetic patients. Islet cell amyloid has also been associated with diabetes. In both types, a variety of viruses have been implicated, most notably congenital rubella, Coxsackievirus B, and cytomegalovirus.


Research has identified two groups of cellular carbohydrate transporters in cell membranes. Sodium-linked glucose transporters are found primarily in the intestine and kidney. The glucose transporter (GLUT) proteins are found throughout the body and transport glucose by facilitated diffusion down concentration gradients. The GLUT-4 transporter, found primarily in muscle, is insulin responsive, and a signaling defect in the protein may be responsible for insulin resistance in some diabetic patients.



Clinical Features




Type 2


The patient with type 2 diabetes is usually middle-aged or older and overweight, with normal to high insulin levels. Insulin levels are lower than would be predicted for glucose levels, however, leading to a relative insulin deficiency. All type 2 patients demonstrate impaired insulin function related to poor insulin production, failure of insulin to reach the site of action, or failure of end-organ response to insulin.


As with type 1 diabetes, research suggests that distinct subgroups of patients fall within the classification of type 2 diabetes. Although most adult patients are obese, 20% are not. Nonobese patients form a subgroup with a different disease, more similar to type 1. Younger patients with type 2 diabetes were previously thought to have a disease with a different course and risk factors, referred to as maturity-onset diabetes of the young. However, one of the fastest-rising subgroups of patients with type 2 diabetes is now young adults and children, who have a disease similar to that seen in older adults and thought to be due to the rise in obesity in this age group.


Symptoms tend to begin more gradually in type 2 diabetes than in type 1. The diagnosis of type 2 is often made by the discovery of an elevated blood glucose level on routine laboratory examination. Hyperglycemia may be controlled by dietary therapy, oral hypoglycemic agents, or insulin administration, depending on the individual. Decompensation of disease usually leads to HHS rather than to ketoacidosis.



Diagnostic Strategies



Serum Glucose


As a rule, any random plasma glucose level above 200 mg/dL, HbA1c value above 6.5%, fasting plasma glucose concentration above 126 mg/dL, or 2-hour postload OGTT is sufficient to establish the diagnosis of diabetes. In the absence of hyperglycemia with metabolic decompensation, these criteria should be confirmed by repeated testing on a different day. Confirmation can be made by the same test or two different tests (fasting plasma glucose and HbA1c, for example). A value of 150 mg/dL is likely to distinguish diabetic from nondiabetic patients more accurately. Formal OGTTs are unnecessary except during pregnancy or in patients who are thought to have diabetes but who do not meet the criteria for a particular classification. The World Health Organization and ADA provide protocols for performance of the OGTT.1



Glycosylated Hemoglobin


Measurement of glycosylated hemoglobin (HbA1c) is one of the most important ways to assess the level of glucose control. Elevated serum glucose binds progressively and irreversibly to the amino-terminal valine of the hemoglobin β chain. The HbA1c measurement provides insight into the quality of glycemic control over time. Given the long half-life of red blood cells, the percentage of HbA1c is an index of glucose concentration of the preceding 6 to 8 weeks, with normal values approximately 4 to 6% of total hemoglobin, depending on the assay used. Levels in patients with poorly controlled disease may reach 10 to 12%. Measurement of glycated albumin can be used to monitor diabetic control during 1 to 2 weeks because of its short half-life but is rarely used clinically. The ADA recommends at least biannual measurements of HbA1c for the follow-up of all types of diabetes. The ADA currently sets an HbA1c value of less than 7% as a treatment goal.



Urine Glucose


Urine glucose measurement methods are basically of two types: reagent tests and dipstick tests. The reagent tests (e.g., Clinitest) are copper reduction tests. They are somewhat more cumbersome and expensive than dipstick methods, use tablets that are caustic and dangerous if accidentally ingested, and may be affected by many substances.


Dipstick tests generally use glucose oxidase, which may also be affected by different substances. Dipsticks are inexpensive and convenient but may vary in their sensitivity and strength of reaction to a given concentration of glucose. Dipstick interpretation can vary significantly, depending on the observer and the type of lighting. Both falsely high and falsely low urine glucose readings can also occur. With the “plus” system, 1+, 2+, 3+, and 4+ have different implications about urine glucose concentrations, depending on the brand of dipstick. The use of reflectance colorimeters to read dipsticks increases accuracy. Urine glucose tests must be interpreted loosely because many factors can affect their results.




Dipstick Blood Glucose


Dipsticks for testing of blood glucose are clearly more accurate than urine dipsticks as a means of monitoring blood glucose concentration, but they also may be inaccurate. Hematocrits below 30% or above 55% cause unduly high or low readings, respectively, and a number of the strips specifically disclaim accuracy when they are used for neonates. Sensitivity of dipsticks to a variety of factors varies with the particular brand. The largest errors are in the hyperglycemic range. Dipstick readings rarely err more than 30 mg/dL when actual concentrations are below 90 mg/dL. Although specific glucose concentrations may not be accurately represented, blood glucose dipsticks are useful in estimating the general range of the glucose value. Reflectance meters increase the accuracy of the dipstick blood glucose determination. If maximum accuracy is desired, a laboratory blood glucose level should be obtained.



Hypoglycemia


Hypoglycemia is a common problem in patients with type 1 diabetes, especially if tight glycemic control is practiced; it may be the most dangerous acute complication of diabetes. The estimated incidence of hypoglycemia in diabetic patients is 9 to 120 episodes per 100 patient-years. As significant efforts continue to keep both fasting and postprandial glucose concentrations within the normal range, the incidence of hypoglycemia may increase. The most common cause of coma associated with diabetes is an excess of administered insulin with respect to glucose intake. Hypoglycemia may be associated with significant morbidity and mortality.2 Severe hypoglycemia is usually associated with a blood glucose level below 40 to 50 mg/dL and impaired cognitive function.



Principles of Disease


Protection against hypoglycemia is normally provided by cessation of insulin release and mobilization of counter-regulatory hormones, which increase hepatic glucose production and decrease glucose use. Diabetic patients using insulin are vulnerable to hypoglycemia because of insulin excess and failure of the counter-regulatory system.


Hypoglycemia has many causes: missing a meal (decreased intake), increased energy output (exercise), and increased insulin dosage. It can also occur in the absence of any precipitant. Oral hypoglycemic agents have also been implicated in causing hypoglycemia, both in the course of therapy and as an agent of overdose.


Hypoglycemia without warning symptoms, or hypoglycemia unawareness, is a dangerous complication of type 1 diabetes probably caused by previous exposure to low blood glucose concentrations.2 Even a single hypoglycemic episode can reduce neurohumoral counter-regulatory responses to subsequent episodes. Other factors associated with recurrent hypoglycemic attacks include overaggressive or intensified insulin therapy, longer history of diabetes, autonomic neuropathy, and decreased epinephrine secretion or sensitivity.


The Somogyi phenomenon is a common problem associated with iatrogenic hypoglycemia in the type 1 diabetic patient. The phenomenon is initiated by an excessive insulin dosage, which results in an unrecognized hypoglycemic episode that usually occurs in the early morning while the patient is sleeping. The counter-regulatory hormone response produces rebound hyperglycemia, evident when the patient awakens. Often, both the patient and the physician interpret this hyperglycemia as an indication to increase the insulin dosage, which exacerbates the problem. Instead, the insulin dosage should be lowered or the timing changed.



Clinical Features


Symptomatic hypoglycemia occurs in most adults at a blood glucose level of 40 to 50 mg/dL. The rate at which glucose decreases, however, and the patient’s age, gender, size, overall health, and previous hypoglycemic reactions contribute to symptom development. Signs and symptoms of hypoglycemia are caused by excessive secretion of epinephrine and CNS dysfunction and include sweating, nervousness, tremor, tachycardia, hunger, and neurologic symptoms ranging from bizarre behavior and confusion to seizures and coma. In patients with hypoglycemia unawareness, the prodrome to marked hypoglycemia may be minimal or absent. These individuals may rapidly become unarousable without warning. They may have a seizure or show focal neurologic signs, which resolve with glucose administration.




Management


In alert patients with mild symptoms, oral consumption of sugar-containing foods or beverages is often adequate. In other patients, after blood is drawn for glucose determination, one to three ampules of 50% dextrose in water (D50W) is administered intravenously while the patient’s airway, breathing, and circulation are assessed and maintained. Augmentation of the blood glucose level by administration of an ampule of D50W may range from less than 40 mg/dL to more than 350 mg/dL. These therapeutic steps are appropriately performed in the field if out-of-hospital care is available. If alcohol abuse is suggested, thiamine is administered. D50W should not be used in infants or young children because venous sclerosis can lead to rebound hypoglycemia. In a child younger than 8 years, it is advisable to use 25% (D25W) or even 10% (D10W) dextrose. D25W may be prepared by diluting D50W 1 : 1 with sterile water. The dose is 0.5 to 1 g/kg body weight or, using D25W, 2 to 4 mL/kg.


If intravenous access cannot be rapidly obtained, 1 to 2 mg of glucagon may be given intramuscularly or subcutaneously. The onset of action is 10 to 20 minutes, and peak response occurs in 30 to 60 minutes. It may be repeated as needed. Glucagon may also be administered intravenously; 1 mg has an effect similar to that of one ampule of D50W. Glucagon is ineffective in causes of hypoglycemia in which glycogen is absent, notably alcohol-induced hypoglycemia.


Families of type 1 diabetic patients are often taught to administer glucagon intramuscularly at home. Of the families so instructed, only 9 to 42% actually inject the glucagon when it is indicated. Intranasal glucagon has not been widely used. Out-of-hospital care providers and emergency physicians should seek a history of glucagon administration because it alters initial blood glucose readings.


All patients with severe hypoglycemic reactions require aspiration and seizure precautions. Although the response to intravenous administration of glucose is generally rapid, older patients may require several days for complete recovery.


Treatment of hypoglycemia secondary to oral hypoglycemic agents depends on the agent. Metformin and the thiazolidinedione agents rarely cause significant or prolonged hypoglycemia, whereas sulfonylureas, which are insulin secretagogues, do cause hypoglycemia. Sulfonylurea oral hypoglycemic agents pose special problems because the hypoglycemia they induce tends to be prolonged and severe. Patients with an overdose of sulfonylurea hypoglycemic agents should have a minimum observation period of 24 hours if hypoglycemia is recurrent in the ED after management of the initial episode. Patients at risk for hypoglycemia from oral sulfonylureas include patients with impaired renal function, pediatric patients, and patients who are naïve to hypoglycemic agents. Although symptoms may occur after an overdose, several case reports in patients with renal failure and pediatric patients describe refractory hypoglycemia after ingestion of a single pill. One case series of pediatric patients presenting with sulfonylurea ingestion who were euglycemic initially demonstrated an average time to onset of 8 hours to the initial hypoglycemic episode.3 However, onset of symptoms was delayed up to 18 hours in some patients. As a result, we recommend 23 hours of observation for patients with known or suspected ingestion of hypoglycemic agents.


A patient with hypoglycemia from sulfonylureas, in addition to standard glucose replacement, frequently requires treatment with an agent to inhibit further insulin release, such as octreotide, a somatostatin analogue. Several case series have described the use of octreotide in both adult and pediatric patients suffering from sulfonylurea-induced hypoglycemia, frequently reporting successful results with a significant decrease in the number of episodes of recurrent hypoglycemia. A randomized clinical trial concluded that patients receiving octreotide had a decreased glucose supplementation requirement.4 No single set protocol for use has been described; however, typical adult doses have ranged from 50 to 100 µg intravenously or subcutaneously every 12 hours, and pediatric dosages have ranged from 25 to 50 µg intravenously or subcutaneously. Whereas experience thus far with octreotide has been positive, it does not obviate the need for prolonged observation and serial glucose measurements.



Disposition


Type 1 diabetic patients with brief episodes of hypoglycemia uncomplicated by other disease may be discharged from the ED if a cause of the hypoglycemia can be identified and corrected by instruction or medication. All patients should be given a meal before discharge to ensure their ability to tolerate oral feedings and to begin to replenish glycogen stores in glycogen-deficient patients. Patients who are discharged should receive short-term follow-up for ongoing evaluation. Patients with hypoglycemia caused by long-acting sulfonylurea medications should be observed in the hospital if they have recurrent hypoglycemia after a period of observation in the ED. Other agents, such as metformin, do not typically produce hypoglycemia, although they may have other issues, such as lactic acidosis, that may require admission.



Nondiabetic Patients


Hypoglycemia in the nondiabetic patient may be classified as postprandial or fasting. The most common cause of postprandial hypoglycemia is alimentary hyperinsulinism, such as that seen in patients who have undergone gastrectomy, gastrojejunostomy, pyloroplasty, or vagotomy. Fasting hypoglycemia is caused when there is an imbalance between glucose production and use. The causes of inadequate glucose production include hormone deficiencies, enzyme defects, substrate deficiencies, severe liver disease, and drugs. Causes of overuse of glucose include the presence of an insulinoma, exogenous insulin, sulfonylureas, drugs, endotoxic shock, extrapancreatic tumors, and a variety of enzyme deficiencies.


Emergency treatment is similar to that of hypoglycemia in the diabetic patient. The determination of inpatient versus outpatient evaluation of hypoglycemia in a nondiabetic patient should be based on the suggested cause and the nature of the episode (i.e., factors such as severity, persistence, and recurrence).

< div class='tao-gold-member'>

Related posts:

Default ThumbnailBlood and Blood Components Brain Resuscitation Gastrointestinal Bleeding Dizziness and Vertigo Multiple Trauma Somatoform Disorders
Tags:
Jul 26, 2016 | Posted by in ANESTHESIA | Comments Off on Diabetes Mellitus and Disorders of Glucose Homeostasis

Full access? Get Clinical Tree
Get Clinical Tree app for offline access