November 1997

VI DIABETES MELLITUS
Diabetes mellitus is a disease of metabolic dysregulation, most notably dysregulation
of glucose metabolism, accompanied by long-term vascular and neurologic
complications. Diabetes has several clinical forms, each of which has a distinct
etiology, clinical presentation, and course. Insight into diabetes and its complications
has been advanced by extensive metabolic studies, the use of radioimmunoassays for
insulin and glucagon, and the application of molecular biology strategies.

The identification of what was once termed isletin (i.e., insulin) as the active
glucose-lowering agent in the islets and its isolation and use in diabetic animals by
Banting and Best in Toronto and by Paulesco in Romania culminated in the
successful treatment of Leonard Thompson on January 11, 1922.1,2 The availability
of insulin cured an otherwise uniformly fatal disease. Further clinical experience,
however, revealed that insulin-treated patients developed seemingly inexorable long-
term complications leading to blindness, kidney failure, and peripheral and
cardiovascular disease. A uniformly fatal disease had thus been transformed into a
chronic disease with significant morbidity and premature mortality.

In the 75 years since its first use, insulin has been purified and its amino acid
sequence established. Insulin has been synthesized by solid-phase techniques and,
more recently, by modification of porcine insulin and by recombinant methods in
bacteria or yeast, providing a limitless supply. Modifications of insulin have
provided an array of insulin formulations that can be used creatively to achieve
specific glucose targets.3 Insulin analogues-created by the substitution of specific
amino acids-which do not self-aggregate and are more rapidly acting, have been
approved, and others are being investigated.4 Whole organ pancreas transplantation
has become increasingly successful in recent years,5 and the transplantation of
isolated islets and the use of hybrid devices (i.e., islets encapsulated in artificial
chambers) are under active investigation.6,7 The Diabetes Control and Complications
Trial (DCCT) demonstrated the effectiveness of intensive therapy in maintaining
normal glucose control and preventing or delaying the long-term complications of
diabetes. DCCT established a new standard of therapy for insulin-dependent diabetes
mellitus (IDDM) and, by extension, for non-insulin-dependent diabetes mellitus
(NIDDM).8 New methods of intensive therapy that are more user-friendly and less
likely to cause hypoglycemia must also be established. The development of
noninvasive glucose sensors9 and implantable insulin pumps10 will advance this
cause. New therapies directed at preventing complications, independent of glycemic
control, are being investigated, and other studies are examining whether IDDM and
NIDDM can be prevented.

Epidemiology
All of the early observations regarding diabetes focused on its most dramatic
expression, IDDM, because it affected children and ran a rapid downhill course.
IDDM and the other major clinical form of diabetes, NIDDM, were not clearly
distinguished until relatively recently. In 1936, the existence of IDDM (then termed
insulin-sensitive diabetes) and NIDDM (then termed insulin-insensitive diabetes)
was first documented.11 Subsequent studies that used bioassays and
radioimmunoassays for insulin demonstrated the pathophysiological difference
between the two forms of diabetes. Patients with IDDM have an absolute insulin
deficiency associated with immunologically mediated destruction of the islets,12
whereas NIDDM patients, who have a nearly normal islet mass, do not secrete
sufficient insulin to meet the increased demand caused by insulin resistance.13

The National Diabetes Data Group (NDDG) and the World Health Organization
(WHO) adopted similar definitions for the different clinical forms of diabetes in
1979.14 The definitions were based on characteristic clinical presentations and
pathophysiology and were supported by immunologic markers and inheritance [see
Table 1]. Although the definitions avoided the age criteria formerly carried by the
terms juvenile-onset and adult-onset diabetes, IDDM is generally a disease of
childhood: most cases appear in patients younger than 20 years, and peak onset is
around the time of puberty. Fewer than 10 percent of IDDM cases may first appear
in patients older than 50 years. Such patients are thin and prone to ketosis, and they
share genetic and immunologic markers with IDDM patients whose disease begins in
youth. Patients who develop IDDM in the setting of polyendocrine deficiency
syndrome type II typically are 30 years of age or older at onset [see Section 3,
Subsection IV]15; most cases of NIDDM occur in patients older than 40 years.
NIDDM develops in a small percentage of patients in their 20s and 30s. Such
patients are often very obese, the offspring of two parents with NIDDM, women with
a history of gestational diabetes, or members of a racial or ethnic group with a
particularly high prevalence of NIDDM [seeNIDDM, below]. The American
Diabetes Association issued revised guidelines for use of fasting plasma glucose
levels as the criterion for diagnosis of diabetes in 1997 [see Table 1]. 16

IDDM

IDDM is rare, affecting one in 250 persons in the United States, where
approximately 10,000 to 15,000 new cases are reported each year [see Table 2]. 17
The highest prevalence of IDDM is found in northern Europe, where more than one
in every 150 Finns develop IDDM by 15 years of age. Data suggest that for unknown
reasons, the incidence of IDDM is increasing in Europe. IDDM is less common
among black and Asian populations, in which the frequency of IDDM is less than
half that among whites.

NIDDM

Compared with IDDM, NIDDM is common [see Table 2], with an overall
prevalence of 6.6 percent in the United States.18 In the second half of the 20th
century, NIDDM has become one of the most frequent chronic diseases in the United
States and most other industrialized nations.19 More than 600,000 new cases are
reported each year, and the projected prevalence for the next decade is 10 percent.
One half of the NIDDM population are unaware of their disorder.18 The increase in
the prevalence of NIDDM in the United States is commonly attributed to an aging
population that is also increasingly obese and sedentary. NIDDM is particularly
common in certain racial and ethnic groups, such as Native Americans; the natives of
Nauru, in the South Pacific, and Mauritius, in the Indian Ocean; Hispanic
Americans; Asians; and African Americans. The prevalence of NIDDM among
persons older than 65 years exceeds 18 percent, and compared with normal-weight
individuals, obese people (i.e., those who have a body mass index greater than 30)
are at 10 to 20 times greater risk for NIDDM.20 Almost 90 percent of NIDDM
patients are obese. Abdominal (i.e., android, as opposed to gynecoid) obesity, which
reflects increased visceral fat and can be conveniently measured as an increased
waist-to-hip ratio, is an even more important risk factor for NIDDM than obesity.21
Finally, certain isolated ethnic populations, such as the Pima Indians of Gila River,
Arizona, and the Nauruan islanders, have NIDDM prevalences as high as 50 percent
in the adult population.22,23 The high prevalence of NIDDM in these isolated
populations has developed in the setting of dietary changes, a rising prevalence of
obesity, and an increasingly sedentary lifestyle (on the reservation in the case of the
Pimas and with growing wealth for the Nauruans).

Although genetic and immunologic markers for IDDM had been identified, the
NDDG found them neither specificenough nor sensitive enough to be used to define
IDDM or to distinguish between IDDM and NIDDM. Similarly, measurements of
the secretion of insulin or C-peptide have not reliably differentiated between IDDM
and NIDDM and are not used to define either disorder. Identification of an antibody
directed against glutamic acid decarboxylase (anti-GAD) as a marker for IDDM24
and standardization of the anti-islet cell antibody (ICA) assay25 offer improved
specificity and sensitivity in identifying persons at risk for IDDM. In addition, the
search for the genetic cause of IDDM has identified several loci on chromosome 6
that may account for the major genetic component of IDDM,26 and the use of
candidate gene and genomic strategies has allowed many potential causes of NIDDM
to be ruled out.

OTHER FORMS OF DIABETES

In addition to the two major clinical forms of diabetes, several other forms of
diabetes that have a relatively clear etiology have been recognized. Diabetes that
results from pancreatic destruction (e.g., from recurrent pancreatitis, pancreatectomy,
or toxins such as the rodenticide Vacor) is termed secondary diabetes, as is diabetes
associated with drugs that increase insulin resistance (e.g., glucocorticoids) or
decrease insulin secretion (e.g., phenytoin or beta-adrenergic blockers). Gestational
diabetes usually develops during the last trimester of pregnancy. The vast majority of
these cases are similar in pathophysiology to NIDDM. Patients become glucose
tolerant after delivery; however, 30 to 50 percent of women with gestational diabetes
redevelop NIDDM within 10 years of the initial diagnosis.27 A relatively rare form of
diabetes, in which the affected individuals are not prone to ketosis and which is
inherited as an autosomal dominant trait, has been identified.28 It most commonly
develops in thin adolescents and is known as maturity-onset diabetes of the young
(MODY). A glucokinase with an abnormally high Km (Michaelis constant) has been
identified in some, but not all, families with MODY and is associated with decreased
insulin secretion.29 Another rare form of diabetes that may be associated with insulin
deficiency or can phenotypically resemble NIDDM and is associated with hearing
loss, other central nervous system abnormalities, and a mutant mitochondrial transfer
RNA with maternal inheritance has been identified predominantly in Japanese
families.30

Although NIDDM, IDDM, and gestational diabetes constitute most of the cases of
diabetes in the United States and other industrialized countries, famine diabetes may
be the most common form in countries with limited food supplies. Malnutrition and
carbohydrate-restricted diets in particular are known to be associated with decreased
insulin secretion.

IMPAIRED GLUCOSE TOLERANCE

The definition of diabetes implies both abnormal glucose levels and long-term
vascular complications, features shared by all forms of diabetes. In the general
population, glucose levels after a glucose challenge (e.g., the 75 g oral glucose
tolerance test) have a normal distribution. Those values that are considered abnormal
but are not associated with specific long-term microvascular complications are
stipulated to represent impaired glucose tolerance. The diagnostic blood glucose
values for impaired glucose tolerance were based on the findings of several large
epidemiological studies [see Table 1]. The rationale for this separate diagnostic
category is that patients with impaired glucose tolerance, although at no risk for
retinopathy or nephropathy, are at greater risk for macrovascular disease than
persons with normal glucose tolerance. In addition, 30 to 50 percent of patients with
impaired glucose tolerance develop NIDDM within 10 years after diagnosis.31,32
Possible intervention at this prediabetic stage to prevent subsequent development of
NIDDM is being studied.

Pathogenesis

IDDM

IDDM is characterized by absolute insulin deficiency, making patients dependent on

exogenous insulin for survival. Although IDDM characteristically has an acute
clinical onset with symptoms of hyperglycemia (including polyuria, polydipsia,
weight loss, blurred vision, or all four) preceding ketoacidosis or diagnosis by only
days or weeks, the natural history of IDDM is now known to include a long
asymptomatic preclinical period.33 During this preclinical period, insulin-producing
beta cells are progressively destroyed [see Figure 1]. The normal pancreas contains
1.0 to 1.5 million islets, and roughly 80 percent of islet cells are insulin-secreting
beta cells. Clinical diabetes appears when more than 90 percent of the islet beta cells
have been destroyed.
 Figure 1

Stages of
IDDM/Beta Cell
Mass

The autoimmune destruction of beta cells is associated with, and probably caused by,
lymphocytic infiltration, termed insulitis. Characteristic changes in T cell subsets (e.
g., increased suppressor-inducer T cells and decreased helper-inducer T cells) have
been observed to precede islet destruction.34 Patients destined to develop IDDM can
be identified by the appearance of ICAs as long as 10 years before the clinical
appearance of diabetes.33 These antibodies are not cytotoxic and are probably
markers of immune destruction. Although several studies have suggested that islet
destruction is inexorably progressive, other data suggest that the presence of ICAs
might be intermittent and that the presence of these antibodies, especially in low
titers, might not reliably predict the development of IDDM.35 Other antibodies,
directed against insulin and glutamic acid decarboxylase, have also been identified
early in the evolution of IDDM.36 When more than one of the antibodies is present in
association with decreased insulin secretion or with an HLA type that is
characteristic of IDDM, the predictive value of the tests increases.37 Several
autoantigens that may be involved in the pathogenesis of IDDM, including insulin,
GAD65, GAD67, heat shock protein, and other islet cell antigens, have been
identified.38 Whether viral infections precipitate or cause IDDM in susceptible
persons is not clear.

Genetics

Although IDDM is clearly a heritable disease, the pattern of heredity has escaped
definition. In a study of monozygotic (i.e., identical) twins and triplets among whom
IDDM had developed in one sibling, the disease developed in only 50 percent of the
remaining siblings.39 This observation strongly suggests that an environmental
influence is superimposed on the heritable component of IDDM. Follow-up of such
pairs and triplets has demonstrated that individuals with IDDM develop ICAs first,
followed by decreased insulin secretion in response to intravenous glucose.33 Fasting
and postprandial plasma glucose levels of such people remain in the normal range
until shortly before the appearance of clinical diabetes.

Some HLA loci (DR3 and DR4) are associated with an increased risk of developing
IDDM, whereas other loci appear to be protective.40 Substitution of alanine, valine,
or serine for the more usual aspartic acid at position 57 of the  chain encoded by the
HLA-DQ locus also has been found to be closely associated with the increased risk
of developing IDDM.41 All of the genetic linkages are located in the MHC class II
region on chromosome 6.

NIDDM

Insulin Resistance in NIDDM

Although the immediate cause of IDDM, the destruction of insulin-producing beta

cells, has been recognized for decades, the cause of NIDDM continues to be debated
vigorously [see Figure 2]. In their first studies of circulating insulin levels measured
by radioimmunoassay, Berson and Yalow noted normal or even high insulin levels in
older, obese, non-insulin-treated diabetic patients.42, 43 Consistent with this finding,
autopsy studies of pancreata from patients who had NIDDM revealed only modestly
reduced islet mass compared with pancreata from nondiabetic individuals of a
similar age and weight.44 To explain the observation of increased glucose levels with
normal or high absolute insulin levels, the concept of insulin resistance (i.e.,
decreased sensitivity to insulin action) was proposed. After the first step in insulin
action was identified as the binding of insulin to specific high-affinity cell surface
insulin receptors, in vivo studies using insulin clamp techniques demonstrated a
modest decrease in insulin receptors in insulin-resistant patients.45 The major locus of
resistance, predominantly in muscle, appears to be distal to the receptor (postrecep
tor). Further identification, isolation, and cloning of the insulin receptor has allowed
the study of its function in nondiabetic individuals and in patients with NIDDM. In
the vast majority of cases of NIDDM, no abnormalities are apparent either in the
structure of the insulin receptor or in the tyrosine kinase that is integral to its
function. However, insulin-mediated stimulation of tyrosine kinase and autophos
phorylation are impaired in NIDDM.46 These effects are reversible with improved
glycemia and probably do not entirely account for insulin resistance. Identification of
a series of substrates for the receptor tyrosine kinase, including insulin receptor
substrates 1 and 2, which, when activated, serve as docking proteins for other
insulin-responsive proteins, has advanced understanding of insulin action.47 No
primary abnormalities in these proteins have been documented in NIDDM.

Figure 2

Mechanisms of
 Normal Glucose
Homeostasis and
Insulin Resistance

The major sequela of insulin resistance is decreased muscle uptake of glucose. This
abnormality would explain the generally higher postprandial glucose values seen in
persons with impaired glucose tolerance or early NIDDM, because the major fraction
of p ostprandial glucose disposal is by muscle uptake.48 paired glucose uptake by
muscle may be secondary to abnormal glucose transporters; decreased glucose
phosphorylation, which is the rate-limiting step in glucose metabolism in the muscle;
impaired glycogen synthesis; relatively unsuppressed glycogenolysis or
gluconeogenesis; or a combination of these conditions. Unfortunately, insulin
deficiency and hyperglycemia can cause many of these conditions, making the
identification of the primary underlying abnormality problematic.

The Wounded Beta Cell in NIDDM

The demonstration of insulin resistance in many, if not all, patients with NIDDM is
not sufficient, however, to explain the evolution of impaired glucose tolerance to
frank diabetes. First, even in the presence of decreased glucose transport and insulin
resistance, the increased hepatic glucose output that causes the fasting hyperglycemia
most characteristic of NIDDM remains unexplained. Second, although many persons
are insulin resistant-including most obese persons and patients with polycystic ovary
syndrome, acromegaly, and Cushing's syndrome-only a minority of these persons
develop NIDDM. All of the defects associated with insulin resistance can be
overcome with increased insulin delivery. Thus, most patients with insulin resistance
do not develop diabetes, because they can increase insulin secretion sufficiently to
overcome the resistance. Patients who decompensate and develop NIDDM either are
incapable of increasing insulin secretion beyond a certain level or seem to tire of
producing increased insulin and develop NIDDM as insulin secretion wanes.13 In
fact, although many studies have demonstrated hyperinsulinemia in patients with
impaired glucose tolerance, a decrease in insulin secretion uniformly accompanies
the transition to NIDDM.49 In addition, although the absolute level of insulin
secretion may be elevated before clinical diabetes develops in the patient, 20 to 40
percent of the insulin is proinsulin and proinsulin split products, which represent
incomplete cleavage of the connecting peptide (or C-peptide) from proinsulin during
posttranslational processing. In contrast, proinsulin accounts for less than 15 percent
of total insulin in nondiabetics. Because proinsulin and its split products are less than
10 percent as biologically active as insulin, the true insulin activity in prediabetes
may not be elevated.50

As insulin secretion wanes, the loss of first-phase insulin secretion is a common

finding.51 Normal insulin secretion in response to a rapidly administered intravenous
glucose dose includes a rapid release of insulin in the first one to three minutes and a
more prolonged release of insulin beginning 10 minutes after infusion [see Figure 3].
A third phase, consisting of a very slow leak of insulin, has also been described. The
first phase, or spike, of the spike-and-dome secretion profile is lost not only in
NIDDM but also by the waning beta cells of the pancreas in IDDM. A similar
phenomenon has been induced in animal models by partial pancreatectomy.52 Thus,
the loss of first-phase secretion is common in limited or stressed beta cell masses.

Figure 3

Insulin Secretion
Profiles

A major feature of NIDDM is that both insulin resistance and insulin secretion are
improved if basal (or fasting) glucose levels are normalized, regardless of the means
of normalization. Dietary, insulin, and oral agent therapies have all been documented
to improve endogenous insulin secretion.53-56 The reversible effect of elevated
glucose concentrations on insulin resistance and secretion has been termed
glucotoxicity. Although its mechanism is unknown, glucotoxicity may explain why
hyperglycemia begets worse hyperglycemia. Finally, a specific pattern of central
obesity, reflecting increased visceral fat deposition, is particularly related to the risk
of developing NIDDM.20 The increased release of free fatty acids from these visceral
fat stores, with direct delivery to the liver and pancreas, has been postulated to
underlie abnormal lipid metabolism and insulin resistance.

Genetics

In view of these observations, abnormal insulin secretion must be considered integral

to the development of NIDDM; it may even be the rate-limiting step. Studies of
monozygotic twins have demonstrated a strong genetic influence in NIDDM;
virtually all identical siblings of NIDDM patients also develop NIDDM.39 Patients at
high risk for NIDDM (e.g., Pima Indians and children of two parents with NIDDM)
are insulin resistant before impaired glucose tolerance develops.57,58 Although several
studies have demonstrated normal or high insulin secretion at this stage, others have
shown abnormal secretory pulsations or decreased secretion.59-61 No HLA markers
specific for NIDDM have been identified.

Thus, NIDDM is an inherited disorder in which insulin resistance and abnormal

insulin secretion play important pathogenetic roles. Impaired glucose tolerance
precedes the development of NIDDM and is characterized by insulin resistance,
which causes postp randial hyperglycemia because of decreased muscle uptake of
glucose. Insulin resistance appears to be a heritable characteristic. Although 30 to 50
percent of patients having impaired glucose tolerance progress to NIDDM, the
remainder do not. The critical factor associated with the progression from impaired
glucose tolerance to NIDDM is a decline in insulin secretion. Whether abnormal
insulin secretion is acquired or also inherited is unknown. However, the possibility
that two independently inherited t raits are involved in a disease as common as
NIDDM seems remote. No mutations specific to insulin secretion or insulin
resistance, using candidate gene and genomic strategies, have been identified to
explain the majority of NIDDM cases. MODY is the only form of NIDDM with a
well-characterized single-gene mutation, which provides a pathophysiologic
explanation for its glucose intolerance (although in only about one half of the
families with MODY); mutations in glucokinase lead to defective insulin secretion.29

SECONDARY DIABETES

Diabetes secondary to other causes is relatively rare and generally easy to diagnose.
Its pathophysiology is also readily understood. Several causes of secondary diabetes,
such as chronic pancreatitis or pancreatectomy, result in a loss of beta cells and
insulin secretory capacity. A large fraction of beta cell mass must be lost and
pancreatic damage must be profound before diabetes develops. Drugs such as
glucocorticoids increase insulin resistance. Gestational diabetes develops in the
setting of elevated placental somatomammotropin (human placental lactogen), which
increases resistance.

Metabolism

IDDM

Insulin is the major anabolic hormone in humans. When insulin requirements and
insulin delivery are mismatched, the consequences are abnormal glucose, lipid, and
protein metabolism. Insulin causes its myriad cellular effects by binding to specific
high-affinity receptors. The insulin receptor has been isolated and purified and its
DNA sequence determined and cloned.62 The insulin receptor is a heterodimer with
an extracellular binding domain, a transmembrane domain, and an intracellular
domain, which has tyrosine kinase activity [see Figure 4]. The tyrosine kinase
moiety promotes autophosphorylation at several sites after insulin binding.
Phosphorylation and dephosphorylation play an important role in mediating the
activity of the insulin receptor. Several proximal substrates of the activated receptor
have been identified, including insulin receptor substrates 1 and 2, which bind and
activate proteins that mediate insulin's biologic effects.47

Figure 4

Insulin Receptor
In IDDM, there is clearly no major abnormality in the insulin receptor or in insulin
action. The predominant abnormality is insulin deficiency, which leads to the
following consequences [see Figure 5a and 5b].

Figure 5a Figure 5b

Anabolic Effects of Catabolic Effects

Insulin of Insulin

1. Hyperglycemia secondary to unrestrained hepatic glucose output, resulting

from glycogenolysis and gluconeogenesis, and decreased muscle uptake.
2. Elevated levels of free fatty acids secondary to lipolysis.

3. Elevated serum ketone levels secondary to unrestrained hepatic ketogenesis

from free fatty acids.

4. Increased levels of triglycerides secondary to decreased lipoprotein lipase,

resulting in increased synthesis of very low density lipoproteins (VLDLs) and
decreased VLDL clearance.

5. An increase in branched-chain amino acids and a decrease in protein

synthesis.

6. Because glucose acts as an osmotic diuretic, hyperglycemia results in

dehydration when glucose levels exceed the renal tubular reabsorption
maximum.

Although insulin deficiency is sufficient to cause all of the abnormalities noted

above, elevated levels of counterregulatory hormones such as glucagon, epinephrine,
norepinephrine, growth hormone, and cortisol exacerbate many of the metabolic
abnormalities.63 In patients who are insulin deficient, elevated levels of glucagon and
catecholamines promote ketogenesis and hepatic glucose output. Diabetic
ketoacidosis can develop in patients without increased levels of glucagon, growth
hormone, or catecholamines, but it develops at a slower rate than in patients with
elevated levels of counterregulatory hormones.64 Because ketogenesis is more
sensitive than hepatic glucose output to insulin suppression, low levels of insulin are
usually associated with hyperglycemia but not with ketosis.
The complete destruction of insulin-producing beta cells in the islets of Langerhans
predisposes IDDM patients to labile glucose control. In the absence of significant
quantities of endogenous insulin, glucose metabolism depends completely on
exogenous i nsulin. The progressive destruction of beta cells that is now recognized
to constitute the natural history of IDDM predictably leads to metabolic instability as
the diabetic lesions advance. Early in the clinical course of IDDM, however,
metabolic instability is not invariably present. The first recognized manifestation of
this phenomenon is the so-called honeymoon period in recent-onset IDDM patients.
During this period, insulin can be tapered or even discontinued for weeks or months.
Presumably, patients in whom the honeymoon period develops still retain sufficient
beta cell function to survive without exogenous insulin. Over time, however, the beta
cell mass declines further, eventually resulting in complete insulin deficiency. In
addition to the honeymoon period, patients in whom IDDM develops at a later age
(i.e., in their late teens or 20s) characteristically have a milder early course. During
this period of stability, insulin secretion can be stimulated, albeit at low levels.
Relatively normal glucose control often can be effected with single-injection insulin
regimens. Stable metabolic control can also be maintained in IDDM patients who are
treated very early in the course of their diabetes with intensive insulin regimens
aimed at maintaining normoglycemia. Such treatment has been demonstrated to
preserve endogenous secretion.65 Although beta cell function wanes over time, it
does so more slowly than in conventionally treated patients.
After an initial period of metabolic stability that is generally short but variable, most
IDDM patients develop the full-blown insulin deficiency and metabolic instability
that characterize IDDM [see Figure 6]. Overall, glucose control is highly variable in
IDDM because of the mismatch between insulin requirements and insulin delivery in
most conventional insulin regimens. Even with intensive insulin regimens that rely
on frequent self-monitoring of blood glucose to guide the dose and timing of insulin
injections, glucose control is not normal. The labile glucose levels can be explained
by several variables that influence glucose control in diabetes, including prandial
state, exercise, stress, and nonphysiologic insulin delivery. In turn, numerous
variables influence insulin availability [see Table 3].

Figure 6

IDDM and NIDDM

Glucose Profiles
The most dramatic expression of metabolic instability in IDDM is ketoacidosis.
Ketoacidosis can be precipitated by interruption of insulin therapy or by a stress
state, with elevated counterregulatory hormone levels, and inadequate insulin
therapy. Insul in deficiency and glucagon enhance ketone production from free fatty
acids supplied by lipolysis. Ketogenesis and unrestrained hepatic glucose output lead
to hyperglycemia with consequent dehydration and ketoacidosis.
NIDDM
The metabolic consequences of partial insulin deficiency (or insulin resistance with
inadequate insulin) are predictable. Hyperglycemia leads to dehydration and a
hyperosmolar state. In contrast to patients who have IDDM, patients who have
NIDDM are ketosis resistant because their ambient insulin concentration is generally
sufficient to suppress ketogenesis. However, under extreme stress (e.g., septic shock
or myocardial infarction), NIDDM patients can develop ketoacidosis. More
commonly, elderly patients with NIDDM are at risk for hyperglycemic,
hyperosmolar, nonketotic coma.66 This metabolic emergency develops in NIDDM
patients who become increasingly hyperglycemic because of stress, steroid
administration, or tube feedings. As glucose levels rise in these patients, failure to
replace free-water loss, often because of impaired thirst or mental status or denial of
access to fluids, leads to increasing hyperglycemia and osmolarity, and, ultimately,
to depressed mentation.
In general, glucose levels in NIDDM are far less labile than in IDDM, tending to be
lower before meals and to increase after meals [see Figure 6]. Elevated postprandial
glucose levels in patients who have impaired glucose tolerance result from decreased
muscle uptake associated with insulin resistance. With the development of NIDDM,
however, increased hepatic glucose output causes fasting hyperglycemia and
significantly contributes to postprandial hyperglycemia.67 Fasting glucose levels and
measures of chronic glycemia tend to be stable and reproducible in a given
unstressed NIDDM patient.68,69 Other metabolic abnormalities commonly associated
with NIDDM include elevated VLDL and triglyceride levels; small, dense low-
density lipoprotein (LDL) particles that are particularly atherogenic; and decreased
levels of high-density lipoprotein (HDL) cholesterol.70
Diagnosis and Screening
The clinical presentation of IDDM should be unmistakable. No special diagnostic
tests other than measurement of blood glucose and, depending on the clinical
presentation, urine and serum ketones and electrolytes are required. Although
screening tests that measure ICAs are commercially available, the sensitivity of such
tests is less than 75 percent, even in ideal circumstances.35 Moreover, a high degree
of specificity is critical for a test of any disease of low prevalence. Combining
measurement of ICAs with insulin response to an intrave nous glucose tolerance test
or the combination of several autoantibodies increases the specificity.37 To be
justifiable, however, a screening program must result in improved health status by
identifying the disease in its pre clinical state. Although the use of prophylactic
immunotherapy is being investigated [see Innovative and Experimental Treatments
of IDDM, below], no such benefit would now accrue, and widespread screening
programs for IDDM are not recommended.
Compared with IDDM patients, patients with NIDDM are much more likely to have
no symptoms or fewer symptoms, and diagnosis may require measurement of fasting
glucose lev els or an oral glucose tolerance test. In 1997, the Expert Committee of
the American Diabetes Association issued new guidelines for the diagnosis of
diabetes. These guidelines were based on epidemiological studies correlating fasting
plasma glucose levels with the prevalence of diabetic retinopathy. The committee
recommends that the fasting plasma glucose level used as the diagnostic criterion for
NIDDM be lowered from 140 mg/dl or greater to 126 mg/dl or greater.16 In addition,
the committee recommends that oral glucose tolerance testing be avoided, except in
cases of suspected gestational diabetes, and that all adults older than 45 years
undergo a fasting glucose test at least once every three years to screen for diabetes
[see Table 1]. Although as many as 50 percent of patients with NIDDM are unaware
of their disorder, the benefits of early diagnosis remain unproved, and widespread
screening with glucose tolerance testing cannot be endorsed.71
management of iddm
The primary component of diabetes care is self-care. Of all chronic diseases, diabetes
is unique because its management involves a host of lifestyle adaptations. In addition
to the medications required, the management of IDDM requires self-monitoring and
attention to all aspects of daily living, from diet to exercise.
GLYCEMIC GOALS OF THERAPY: THE DCCT
The specific goals of therapy formerly included maintaining normal growth,
development, and body weight; preventing symptomatic hyperglycemia, clinical
hypoglycemia, and ketoacidosis; and promptly detecting and treating complications.
The glycemic goals of therapy were highly controversial until the results of the
DCCT demonstrated that intensive therapy directed at maintaining normal glucose
levels had a large and beneficial effect on the development and progression of long-
term diabetic complications.8
The debate regarding the relation between metabolic control and the genesis of long-
term complications in diabetes mellitus began within 10 years of the introduction of
insulin therapy.72-74 During the first decades of insulin use, clinicians began to
observe the occurrence of microvascular and neurologic complications.75,76 These
long-term complications, including retinopa thy, neuropathy, and nephropathy, had
not been observed in the preinsulin era, because patients with IDDM had not
survived long enough for them to occur. Although many theories were advanced to
explain the occurrence of these complications, the most popular hypothesis
suggested imperfect glucose control as the cause. The so-called glucose hypothesis
gained support from studies in animal models of diabetes and from epidemiological
studies.69, 77,78 Although vociferous proponents of the glucose hypothesis espoused
the importance of what they termed tight metabolic control, there were no definitive
clinical data to support the efficacy of any particular mode of therapy in preventing
the development or slowing the progression of complications. Moreover, there were
no available therapies that could maintain normal or near-normal levels of glycemia
over time.
The tools to achieve long-term, near-normal glucose control, such as self-monitoring
of blood glucose and intensive insulin regimens, became available in the late 1970s.79
Objective methods of measuring long-term glycemia and complications were
developed at the same time.80 These developments made it possible to design and
conduct clinical experiments to settle the debate.
Design of the trial
The DCCT, initiated in 1983, was designed to determine whether intensive diabetes
management aimed at achieving glycemic levels as close as possible to the near-
diabetic range would affect the development (primary prevention) or the progression
(secondary intervention) of long-term complications in IDDM.8
Two cohorts of patients were enrolled in the study. At baseline, patients in the
primary prevention cohort were 13 to 39 years of age, had had diabetes for one to
five years, and had no evidence of retinopathy or nephropathy. Patients in the
secondary in tervention cohort had had IDDM for as long as 15 years and had at least
one microaneurysm and an albumin excretion rate of less than 200 mg/24 hr at
baseline.
Intensive treatment and metabolic goals in the trial
The two study cohorts were randomly assigned either to conventional therapy, which
consisted of one or two daily in jections of insulin and daily glucose monitoring, or
to intensive therapy. The clinical goal of conventional therapy was to prevent
symptoms of hyperglycemia or hypoglycemia; however, numerical blood glucose
targets were not specified. The goals of intensive therapy were to achieve blood
glucose control that was as close as possible to the nondiabetic range, including
preprandial blood glucose levels of 70 to 120 mg/dl and peak postprandial levels of
less than 180 mg/dl, and to achieve hemo globin A1c (HbA1c) levels in the
nondiabetic range (< 6.05 percent). To reach these goals, patients who were assigned
to intensive therapy self-administered three or more insulin injections daily or made
use of an insulin pump; patients were guided by frequent self-monitoring of blood
glucose [see Table 4]. Insulin therapy was frequently adjusted in response to changes
in ambient blood glucose levels, diet, and exercise.
Results of the Trial
During the mean follow-up of 6.5 years (range, three to nine years), intensive
therapy produced mean HbA1c levels that were approximately two percent lower than
those produced by conventional treatment (seven percent and nine percent,
respectively); however, intensive therapy did not lower mean HbA1c levels to the
nondiabetic range.8 Compared with conventional therapy, intensive therapy
decreased the risk of development and progression of diabetic retinopathy,
nephropathy, and neuropathy by approximately 40 to 76 percent. The beneficial
effects of intensive therapy appeared to increase over time.
Accompanying the salutary effects of intensive therapy on long-term complications
was an approximately threefold increase in severe hypoglycemia, which was defined
as any episode that required assistance to treat. The incidence of hypoglyce mia that
resulted in coma or seizures or that required emergency room treatment also
increased by approximately twofold to threefold. The more severe hypoglycemic
reactions, such as those resulting in seizures or coma, were relatively rare (16
episodes per 100 patient-years in the intensive treatment group and five episodes per
100 patient-years in the conventional treatment group). Other adverse effects that
accompanied intensive therapy included an increased risk of weight gain and of
catheter infections in the patients who used an insulin pump. In concert, none of
these adverse effects caused significant morbidity or mortality. No macrovascular
events or deaths ascribed to hypoglycemia occurred. Moreover, the increased
frequency of hypoglycemia with intensive therapy had no adverse effects on
neurocognitive function as measured by frequent testing. Finally, despite the
demands of intensive therapy, quality of life did not differ between the two treatment
groups.8
Treatment recommendations
The DCCT Research Group, the American Diabetes Association, and other groups
have concluded that the large and consistent effects of intensive therapy in reducing
the risk of developing long-term complications outweigh the considerable effort and
costs involved in such treatment and the increased risk of hypoglycemia that
accompanies it. Intensive therapy has been en dorsed as the treatment of choice for
most patients with IDDM. Realistically, not all patients will be able to implement
intensive therapy; however, the goal for all patients should be to decrease glycemia
to as close to the normal range as safety and lifestyle considerations will allow. For
patients who cannot implement intensive treatment, therapies of variable intensity
are available [see Table 4].
For more than a decade, intensive therapy has been mandatory for women with
IDDM who are trying to conceive or who are pregnant, because such treatment has
been shown to im prove neonatal outcome.81-83 Intensive management of the pregnant
patient with IDDM is particularly difficult because of changing insulin requirements
during pregnancy and should be attempted only by physicians expert in intensive
therapy.
Many questions and challenges remain in the wake of the DCCT. First, whether and
to what extent nonresearch patients with IDDM can implement intensive therapy
remain to be seen. Second, remarkable resources were made available to facilitate
the implement ation of intensive therapy. Whether satisfactory resources are or can
be made available in the nonresearch setting is unknown. Expert treatment teams
must be established to help implement and manage intensive therapy. Financial
barriers must be removed to make intensive therapy as widely available as possible.
Third, according to a strict interpretation of the study results, it was intensive
therapy, not the achievement of normoglycemia, that proved to be effective in
preventing the development and slowing the progression of diabetic complications. It
is not clear that all methods of treatment designed to achieve normoglycemia have
similar effects. It is likely that the achievement of normoglycemia was the major
reason that intensive therapy was effect ive, and similar treatment methods, such as
the use of artificial pancreata (implantable pumps with built-in glucose sensors), are
likely to provide similar benefits when they become available.
On the other hand, pancreas transplantation, which is another means of achieving
normoglycemia, introduces variables, such as immunosuppression, that might have
effects on outcome. Such therapy may need to be evaluated to determine its cost-
benefit ratio. Finally, no patients with NIDDM, the most common form of diabetes,
were included in the DCCT; therefore, whether the DCCT results can be applied to
patients with NIDDM is unknown.84 The similarity between the complications of
IDDM and those of NIDDM, as well as the similar relation between progression of
retinopathy and levels of chronic glycemia in both diseases, suggests that intensive
therapy should also be effective in NIDDM. However, there are scant direct data to
support this hypothesis. Moreover, no long-term studies have established the relative
risks of different therapies designed to achieve normoglycemia in NIDDM. The high
insulin doses that are ultimately required to treat many patients with NIDDM may be
atherogenic, a concern that tempers enthusiasm for the widespread implementation
of intensive therapy in patients with this disease.
SELF-MONITORING
During the past decade, interest in more accurate self-monitoring of metabolic status
has increased, and more devices to help patients perform such monitoring have been
developed. Patients should perform self-monitoring of blood glucose levels for the
following reasons:

1. Hypoglycemia is the major side effect of insulin therapy. Accurate, timely

self-monitoring of blood glucose is necessary to recognize, prevent, and treat
hypoglycemia.
2. Development of insulin delivery regimens that more closely resemble normal
physiology depends on frequent feedback of ambient glucose levels, which
only self-monitoring can provide.

3. The results of self-monitoring provide information that helps educate diabetic

patients about their disease and the effect of exercise, dietary variation,
illness, and other factors on symptoms and glucose control.

Glucose profiles in IDDM are highly labile, and urine glucose tests cannot provide
accurate information regarding contemporaneous blood glucose levels.85
Furthermore, the consequences of hypoglycemia are potentially severe. Therefore,
the need for an accurate, user-friendly method for frequently measuring blood
glucose levels at home, in school, or on the job is clear. Reagent strips and meters to
make the measurements more precise and accurate have been developed.86
The frequency of monitoring in IDDM depends on the risk of hypoglycemia and the
goals of therapy for a particular individual. At a minimum, IDDM patients should
able to monitor their blood glucose level when they change their insulin doses, diet,
or activity or when hypoglycemia would be particularly dangerous (e.g., before a car
trip). In addition, self-monitoring during intercurrent illnesses can help patients
adjust insulin doses. As the complexity of treatment regimens increases, patients
must check their glucose level more frequently, depending on their specific insulin
regimen and daily activity and meal schedule. Fasting and predinner or prebed
measurements are often employed. Intensive regimens require, at a minimum,
preprandial and prebed measurements. Insulin doses are adjusted according to the
results.
When unusually high blood glucose levels reveal the risk of ketoacidosis resulting
from a superimposed illness, inadequate insulin intake, or both, the patients should
test their urine for ketones and be instructed to call their health care provider if more
than trace urine ketones are found. Ketone testing should also be performed
whenever a patient is nauseated or vomiting, because these symptoms may indicate
ketoacidosis.
MONITORING BY A HEALTH CARE PROVIDER
The utility of measuring the concentration of glycosylated proteins-the nonenzymatic
product of glucose and the amino groups of circulating and membrane proteins-as an
index of chronic glucose levels is well established.80 The concentration of a
glycosylated protein reveals the mean blood glucose concentration integrated over
the lifespan of the protein, which reaches steady state at approximately one half of
the lifespan [see Figure 7]. Thus, the concentration of glycosylated hemoglobin-
HbA1, HbA1c, or glycohemoglobin, depending on the specific method used-indicates
mean glucose levels over approximately 60 days because the average erythrocyte life
span is 120 days; the concentration of a glycosylated albumin or serum protein (e.g.,
fructosamine) reveals mean blood glucose levels over a 20-day period. These
measures provide an objective means of assessing long-term glycemia and judging
the efficacy of therapeutic interventions. In at least one study, patients whose HbA1c
was measured had a better health status, with lower glycemia and fewer
hospitalizations, than a randomly selected group of patients whose HbA1c level was
not made known to the patient or physician.87 In IDDM, three to four HbA1c
measurements a year are sufficient for clinical management. Assays that reflect a
shorter period of glycemia, such as assays of fructosamine, need to be repeated more
frequently and are less useful in diabetes management.

Figure 7

Hemoglobin/Glucose
Levels
INSULIN
Insulin is available in various species, formulations, and degrees of purification. In
general, all insulins available in the United States are highly purified, containing less
than 50 ppm of proinsulin. Porcine and bovine insulins differ from human insulin by
one and three amino acids, respectively. Despite the differences in amino acid
sequence and the anti-insulin antibodies generated when bovine, porcine, or a
combination of bovine and porcine insulin is administered, these insulins exhibit
potency similar to that of human insulin and seldom cause immune-mediated
problems. During the rare cases of antibody-mediated insulin resistance or of local or
systemic insulin allergy, either desensitization or a change to highly purified porcine
or human insulin is effective. Highly purified porcine insulin, containing less than 10
ppm of proinsulin, or human insulin is necessary only in those relatively rare patients
who have immune-mediated complications after receiving less pure, more antigenic
animal-source insulins. In addition, those patients using insulin intermittently (e.g.,
patients with gestational diabetes) and who may be at greater risk for immune-
mediated complications should be treated with pure porcine or human insulin.
Human insulin is synthesized by modifying porcine insulin or by using recombinant
methods in Escherichia coli or yeast. Although relatively few diabetic patients have
an absolute indication for human insulin (or an absolute contraindication for animal-
species insulins), insulin manufacturers would like to shift production from animal to
human insulin for economic reasons.
Insulin formulations may be classified as rapid acting (i.e., regular or crystalline zinc
insulin [CZI]), intermediate acting (i.e., Lente or neutral protamine Hagedorn [NPH]
insulin), or long acting (i.e., Ultralente insulin). A new rapid-acting, recombinant
human insulin (lispro or Humalog), created by the reversal of the 28th and 29th
amino acids of the b chain, does not polymerize as readily as CZI. Instead of
hexamers, most of the insulin is in the monomeric form and is absorbed more rapidly
and cleared more rapidly.4 Although each formulation has a different profile of
action [see Table 5], variability in insulin absorption from person to person, and even
in a single individual, renders these profiles crude estimates. In addition, the human
insulins generally have a slightly faster onset and shorter duration than
corresponding animal-species insulins. Several fixed-combination formulations (e.g.,
70 percent NPH and 30 percent regular) are available. These are only rarely useful in
IDDM because fixed-ratio formulations rarely meet the ever-changing insulin
requirements of patients with IDDM; however, the fixed-combination insulins may
be appropriate in NIDDM. Formulations can be combined in a daily regimen to
provide insulin delivery that approximates normal physiology [see Figure 8]. The
more complex regimens require more frequent injections and glucose tests to guide
the choice of dose. Patients treated with intensive regimens should be given
individual algorithms to help them select the correct dose of regular insulin for a
given blood glucose level, meal, and exercise routine. Because the onset of action of
subcutaneous regular insulin is delayed, however, patients must administer
preprandial regular insulin 30 to 60 minutes before the meal. In this respect, lispro
insulin is more convenient because it can be given 10 to 15 minutes before the meal.

Figure 8

Insulin Profiles for

Six Different
Treatment
Regimens
Insulin delivery devices have been developed to facilitate the frequent administration
of regular insulin that is part of intensive treatment regimens. Mechanical syringes,
air jet injectors, battery-powered external pumps, and implantable pumps have all
been effective in intensive regimens and can result in near-normal glycemia.8, 10, 88
Similar metabolic results can be obtained with three or more daily injections.89,90 The
major risk of most intensive regimens is a twofold to threefold increase in severe
hypoglycemia.91 (A study using implantable pumps that deliver insulin intravenously
or intraperitoneally suggested that in such regimens, severe hypoglycemia may not
be as frequent a problem as it is in subcutaneous regimens.92) In addition, weight
gain often accompanies intensive therapy, and the devices that use subcutaneous
catheters can result in inflammation or infections at the catheter insertion sites.
DIET
Dietary therapy is a crucial component of diabetes care. As with monitoring and
insulin therapy, the level of therapeutic intensity should be based on the patient's
lifestyle and motivation and on the overall goals of diabetes care [see Table 4]. The
modern approach to dietary therapy for IDDM emphasizes matching insulin to diet,
not vice versa. To facilitate matching insulin doses to meals and to prevent
hypoglycemia, the IDDM patient should eat consistent, regular meals. Observation
of a prudent diet that is high in carbohydrates (45 to 50 percent of calories), low in
fat (< 30 percent of calories), and low in cholesterol (< 300 mg) is recommended.93
More sophisticated dietary approaches include exchange lists to help patients
maintain a similar carbohydrate content from meal to meal and optimize the insulin
choice. Because different carbohydrates are digested at different rates and have
different effects on glucose levels, glycemic indices have been developed for use in
more intensive regimens. These indices account for the effects that different
carbohydrate-containing foods have on glucose levels.94 Given the importance of diet
in the management of IDDM, a dietitian should be part of the health care team.
EXERCISE
Another component of diabetes care, exercise, is important in maintaining
cardiovascular conditioning, insulin sensitivity, and general well-being. However,
patients must be instructed to adjust their meals, their insulin intake, or both to
prevent hypoglycemia during, immediately after, or even six to 12 hours after
exercise [see Hypoglycemia, below]. High-impact sports can be particularly
hazardous to patients with advanced retinopathy or to patients with peripheral
neuropathy or vascular disease, who are at risk for foot trauma.
DIABETIC KETOACIDOSIS
Diabetic ketoacidosis (DKA) occurs infrequently, with 10 to 15 cases occurring for
every 1,000 patient-years. Most cases occur after diabetes has been diagnosed;
ketoacidosis is not usually the first manifestation of diabetes.95 With the availability
of self-monitoring of blood glucose and urine ketone monitoring, most patients
should be able to recognize and control evolving DKA; early and aggressive
treatment with extra insulin and fluids can forestall its development. In a sense, DKA
often represents either a failure of patient education or an inability of the patient to
comply with a prescribed regimen.
Although mortality associated with DKA should approach zero, a survey in Rhode
Island revealed a fatality rate of almost 10 percent.95 Despite this sobering statistic,
DKA should be considered a treatable, nonfatal metabolic emergency, and a careful
clinical review should accompany any case that results in death. Fatal cases most
commonly occur when DKA is not recognized as the clinical presentation of IDDM
in children or when it is precipitated by severe illness, such as sepsis or myocardial
infarction, in adults with IDDM. In children, fatal cerebral edema can occur, albeit
rarely; it usually develops six to 12 hours after treatment for DKA is initiated. The
clinical presentation of DKA includes signs and symptoms of dehydration
(secondary to osmotic diuresis and vomiting), Kussmaul's respirations (to increase
expiration of volatile acids), nausea and vomiting, and depressed mentation [see
Table 6]. Plasma glucose is elevated, usually from 350 to 700 mg/dl, and an anion
gap acidosis is typically present, accompanied by elevated levels of ketone bodies
(acetoacetate, -hydroxybutyrate, and acetone) detectable in urine and serum.
Although potassium and phosphate levels may initially be normal or even high,
hydration and correction of the acidosis invariably lowers serum levels, unmasking a
profound total body depletion. Other factors can alter the usual laboratory values of
the disorder. Preliminary insulin treatment by the patient, profound inanition, or
pregnancy can result in so-called euglycemic DKA, in which glucose levels are less
than 250 mg/dl.
The fundamental treatment of DKA includes replacement of lost volume and
electrolytes (especially potassium), insulin administration, and diagnosis and
treatment of underlying medical conditions [see Table 7]. Since its introduction in
1974, so-called low-dose intravenous insulin therapy has remained the most
commonly used treatment for DKA.96 Although treatment in an intensive care unit is
not mandatory, the clinical setting should permit insulin administration and frequent
monitoring of vital signs, glucose, and electrolytes. When the patient is obtunded or
in a coma, his or her airway should be protected to prevent aspiration. A large-
caliber intravenous line is required for the administration of intravenous fluid. A 70
kg patient with DKA has a mean volume deficit of 5 to 8 L, a mean sodium chloride
deficit of 350 to 600 mEq, and a mean potassium deficit of 200 to 400 mEq.
Intravenous replacement of these deficits is critical. As the anion gap is corrected, a
hyperchloremic metabolic acidosis often supervenes.97
Although DKA can be managed satisfactorily with insulin given intramuscularly or
even subcutaneously, intravenous administration is far more predictable and results
in fewer instances of hypoglycemia and hypokalemia. The role of bicarbonate and
phosphate replacement continues to be controversial. No studies have provided
support for bicarbonate use,98 although most textbooks continue to advise
bicarbonate administration when the arterial pH is less than 7.05. Severe phosphate
depletion (a level of < 2.0 mEq/L) can result in rhabdomyolysis and diaphragmatic
abnormalities. It can also lead to 2,3-diphosphoglycerate depletion and thus to
abnormal dissociation of hemoglobin and oxygen. Although phosphate
administration can reverse all of these abnormalities, the importance of phosphate
replacement therapy is questionable; controlled clinical trials have not demonstrated
that phosphate replacement improves outcome in patients with DKA.99
HYPOGLYCEMIA
Hypoglycemia is more common than diabetic ketoacidosis and potentially as
dangerous a metabolic emergency. Clinical hypoglycemia can range from annoying
symptoms accompanying a biochemically low glucose level (< 55 mg/dl) to
confusion, seizures, or loss of consciousness (during severe hypoglycemia). The
symptoms that most commonly characterize nonsevere hypoglycemia are adrenergic
and include tremulousness, diaphoresis, and tachycardia. Obviously, severe
hypoglycemia can have disastrous consequences, depending on the setting.
Glucagon and epinephrine are the major counterregulatory hormones that are
secreted in response to hypoglycemia and restore glucose levels by stimulating
hepatic glucose output as insulin levels decline.100 Furthermore, catecholamine
secretion, which usually accompanies hypoglycemia, can alert the patient to treat the
episode because it produces sympathoadrenal symptoms. Although they are less
important, cortisol, growth hormone, and thyroxine also play a role in maintaining
glucose levels. Patients rapidly recognize most episodes and can effectively treat
themselves with a quickly absorbed oral carbohydrate. Approximately 15 g of
carbohydrate is sufficient to restore blood glucose levels to normal [see Table 8].
The use of complex carbohydrates and foods with a high fat content, such as
chocolate, may delay digestion and absorption of saccharides and is not appropriate
for treatment of hypoglycemia. If a patient cannot swallow or cooperate, either
intravenous dextrose or intramuscular or subcutaneous glucagon should be given.
The administration of 25 ml (i.e., one-half ampule) of 50 percent dextrose (12.5 g) as
an intravenous push is generally sufficient to reverse clinical hypoglycemia. Even
less is required in infants and children. Glucagon injection kits contain 1 mg of
glucagon, which increases glucose levels in 15 to 30 minutes. Glucagon may cause
nausea, vomiting, and headache, especially in children.
Patients with severe hypoglycemia usually respond rapidly to treatment, although
patients who are postictal or in a prolonged coma may require days, or even longer,
to regain normal mental status and cognitive function. In rare instances, neurologic
deficits can be permanent after hypoglycemia. In view of the potential consequences
of prolonged hypoglycemia, hypoglycemia should always be treated immediately.
Patients should be instructed to treat themselves as though they have hypoglycemia
when they suspect it on clinical grounds, even if they are unable to confirm it with a
glucose test.
One purpose for self-monitoring of blood glucose is to aid patients in detecting and
treating hypoglycemia, although the benefit of self-monitoring in reducing
hypoglycemia has not been demonstrated. The consensus, however, is that patients at
risk for hypoglycemia should self-monitor their blood glucose to help adjust their
regimen and prevent hypoglycemia.101
Several factors contribute to the development of severe hypoglycemia [see Table 9].
Hypoglycemic episodes that occur during stress, exercise, or sleep or in the setting of
alcohol or illicit drug use may not result in recognizable hypoglycemic symptoms.
Thus, hypoglycemia may go untreated, leading to more severe hypoglycemia.
Furthermore, some risk factors for hypoglycemia, such as alcohol use, have multiple
effects that can precipitate, prolong, or worsen the severity of hypoglycemia.
Alcohol, for instance, impairs judgment and inhibits hepatic glucose output (thereby
delaying recovery). Patients who do recognize incipient hypoglycemia but who do
not respond expeditiously (e.g., by waiting for a meal in a restaurant or continuing to
drive after symptoms first appear) are also at increased risk for severe hypoglycemia.
Finally, because glucagon and epinephrine are the major defenses against prolonged
hypoglycemia, their absence promotes longer, more severe hypoglycemic episodes
via two mechanisms: (1) without glucagon and epinephrine, hepatic glucose output
in response to hypoglycemia is decreased, and (2) in the absence of epinephrine, the
character of hypoglycemic symptoms may change, resulting in failure to recognize
the symptoms and to treat the episode. The glucagon response to hypoglycemia often
wanes after five years of IDDM.100, 102 the absence of the glucagon response,
epinephrine secretion provides adequate counterregulation; however, epinephrine
response can also be lost, sometimes in association with other autonomic
neuropathies. Overall, many patients lose the ability to counterregulate within 10
years of IDDM onset. Moreover, a lowered glucose threshold for glucagon and
epinephrine release in response to hypoglycemia has been observed in patients
undergoing intensive insulin therapy.103 The lower glucose level necessary to
stimulate counterregulation narrows the safety margin of therapy. For instance, the
first symptom of hypoglycemia may occur only at glucose levels as low as 35 mg/dl
and may consist of confusion, which interferes with self-treatment. The predilection
for hypoglycemia with intensive therapy may be secondary to previous
hypoglycemia; a period of hypoglycemia-free therapy may restore hypoglycemia
awareness, if not normal hormonal counterregulation.104
INNOVATIVE AND EXPERIMENTAL TREATMENTS OF IDDM
As our understanding of the pathophysiology of IDDM has evolved, experimental
treatments for the disorder have also evolved. Although a true artificial pancreas has
not been developed for lack of a reliable glucose sensor to provide automatic
feedback and adjustment of insulin delivery, several implantable pumps are being
investigated.10 These devices currently rely on self-monitored blood glucose levels to
direct the choice of insulin doses.
Whole organ pancreas transplantation is now far more successful in restoring
normoglycemia than in the past but requires major surgery and immunosuppression.
6
For that reason, most transplant programs reserve the procedure for IDDM patients
who are undergoing concurrent kidney transplantation and thus who already require
surgery and immunosuppression. Although complications of diabetes are more likely
to be preventable in younger patients, whole organ transplantation is not appropriate
for them because of the risks associated with major surgery and long-term
immunosuppression.
Given the limitations of whole organ transplantation, investigators have turned to the
transplantation of isolated islets, which does not require major surgery. Furthermore,
the ability to immunomodulate the isolated islets (by masking or removing cell
surface antigens) may allow transplantation with little or no immunosuppression.105
Alternatively, islets can be placed in semipermeable hollow tubes that allow glucose
and insulin to enter and leave but shield the islets from lymphocytes and
macrophages.7 Islet transplants have been successful in animal models and in a very
limited number of human subjects.105,106
Finally, because the immunopathogenesis of IDDM is now better understood and
because IDDM can be detected in its preclinical state, it is theoretically possible to
interrupt IDDM development with immunotherapy. Several studies have successfully
used immunosuppressive drugs, including azathioprine, prednisone, and
cyclosporine, to prevent further islet destruction in IDDM patients identified during
their early clinical presentation. Unfortunately, most of the identified patients had
already lost most islet function, and withdrawing immunotherapy even briefly led to
rapid deterioration.107 Moreover, the toxicity of the drugs makes long-term treatment
difficult. Nevertheless, earlier identification of preclinical IDDM and development of
more specific, less toxic immunotherapy strategies hold the promise of preventing
IDDM. A large, multicenter trial is examining whether insulin therapy, administered
by subcutaneous injection and intravenously or orally, will delay the development of
IDDM in persons identified in the preclinical stage.
Management of NIDDM
The clinical goals of therapy for NIDDM are similar to those for IDDM. Just as the
established costs and risks of therapy and the long-term benefits, as documented in
the DCCT, must be balanced for IDDM, they must also be weighed for NIDDM.
Unfortunately, the long-term benefits and risks of tight control in NIDDM are not as
well studied as in IDDM. A relatively small DCCT-like study in patients with
NIDDM has demonstrated a benefit of intensive therapy on long-term
complications.108 The study was in Japanese patients who were very thin and
required very small doses of insulin, making extrapolation to the non-Japanese
NIDDM population problematic. Although the benefits of intensive therapy in
NIDDM probably mirror those in IDDM, the specific costs and risks of intensive
therapy for NIDDM almost certainly differ from those of IDDM. For example, many
NIDDM patients can be treated with diet, a relatively nonintrusive, cost-effective
treatment, which, when successful, decreases multiple cardiovascular risk factors in
addition to improving glucose tolerance. When normoglycemia can be achieved at
low cost and risk, it should be pursued. As more complex therapies are required to
achieve normoglycemia in NIDDM, the cost-benefit ratio must be considered more
carefully, and as in IDDM, attempts to achieve normoglycemia must be tempered by
individual patient factors. Although in general, normoglycemia in NIDDM can be
achieved with less complex drug regimens and with a lower risk of hypoglycemia
than in IDDM, most NIDDM patients are not treated intensively, exposing them to
the risk of long-term complications.
DIET
Diet is preeminent in NIDDM therapy. No drug therapy can match the extraordinary
risk-to-benefit ratio provided by diet therapy. For the more than 85 percent of
NIDDM patients who are obese, weight loss is the major goal of diet therapy. Given
the increased risk of coronary vascular disease in NIDDM, a prudent low-fat, low-
cholesterol diet is also important. The hypocaloric diet should be structured around
three meals a day and should have realistic goals. Attaining ideal body weight is
often not possible, but fortunately, a modest weight loss (e.g., 5 to 10 kg) may
ameliorate or cure the diabetes.109 Although the many diets that stress liquid
supplements, grapefruit extracts, special vitamin formulations, and celebrity sponsors
may be attractive to patients, there is no evidence that these usually costly
approaches are more successful than diets that stress smaller portions of balanced
meals. In fact, although any hypocaloric diet will result in weight loss, the transition
from supplements and formulations to normal meals often results in significant
weight regain. Protein-sparing, modified fasts, especially those using low-quality
liquid protein supplements (usually extracted from horse by-products), can result in
arrhythmias and sudden death. Very low calorie diets (< 800 kcal/day) effectively
promote weight loss in the very obese but can also be accompanied by significant
complications and must be monitored closely by a knowledgeable physician.
Surgical procedures, such as gastric plication (stapling) with small bowel bypass, can
be very effective, but they can be accompanied by morbidity and mortality and are
therefore reserved for patients with severe obesity.110
The expected benefits from weight loss include decreased hyperglycemia (or
normoglycemia), decreased blood pressure and lipid levels, and decreased risk of
cardiovascular and joint diseases. These benefits usually accrue at no cost or risk to
the patient. The problem, of course, is that the benefits are rarely seen, because the
vast majority of obese patients who diet succeed initially but regain the lost weight
within one to two years.111 Each successive attempt to lose weight is increasingly
difficult. Although diabetic patients have even more to gain by losing weight than do
nondiabetic persons, they appear to have more difficulty doing so than do
nondiabetic individuals.112
MONITORING
Monitoring patients with NIDDM is less demanding than monitoring patients with
IDDM. Glucose profiles of NIDDM patients are relatively stable, and the risk of
hypoglycemia among NIDDM patients is low because at least one third of such
patients are treated with diet alone. In addition, compared with IDDM patients,
NIDDM patients treated with oral agents or with insulin tend to be at lower risk for
hypoglycemic episodes. Finally, the risk of ketoacidosis is extremely low, and urine
testing for ketones is not routinely required. Insulin and sulfonylurea-treated NIDDM
patients who are at risk for hypoglycemia should monitor their blood glucose level,
especially when drug treatment is initiated or doses are being adjusted. After a
satisfactory dose has been determined, the stability of the blood glucose profile
makes frequent testing often unnecessary. Some NIDDM patients are treated with
more complex insulin regimens (e.g., two mixed or split injections) and require more
frequent self-monitoring [see Table 4]. In contrast, NIDDM patients treated with diet
alone do not need to monitor their blood glucose routinely.113
In NIDDM, the correlations between day-to-day fasting glucose levels, between
fasting glucose levels and HbA1c levels, and between HbA1c levels over time are
relatively high (r > 0.80).68,69 This glycemic stability makes monitoring in patients
with NIDDM easier than in patients with IDDM.114 Laboratory measurements of
fasting glucose or HbA1c every three to six months usually provide an accurate index
of chronic glycemia in stable NIDDM patients. If there is any change in diet,
exercise, drug regimens, or health status, more frequent monitoring is necessary.
HYPOGLYCEMIC MEDICATIONS
Hypoglycemic medications may be necessary in NIDDM patients in whom the
dietary approach fails. Diet failure usually means that the patient has not been able to
achieve or maintain adequate weight loss to alleviate the diabetes. Before drug
therapy begins, the risks and costs of the drugs must be balanced against the benefits.
Although the benefits of intensive therapy in NIDDM may be debatable, certain
clinical criteria, such as hyperglycemic symptoms uncontrolled by diet or
hyperglycemia severe enough to lead to dehydration, warrant the use of drug therapy
[see Table 10]. All of the drug therapies are more effective in the setting of
continued attention to diet.
Once the health care provider has discussed the need for medication with the patient,
there are several choices of medication, each with distinct advantages and
disadvantages [see Table 11].
Sulfonylureas are oral medications that act primarily by stimulating endogenous
insulin secretion.115 Their glucose-lowering effects were first noted as a side effect
during the investigation of a sulfonamide antibiotic. Although sulfonyl ureas
decrease insulin resistance, this effect probably plays a minimal role, if any, in
lowering glucose levels. On average, sulfonylureas will lower HbA1c levels by one to
two percent. The sulfonylureas have no effect on and no role in the treatment of
IDDM, and they are usually not effective in nonobese NIDDM patients. Several
sulfonyl ureas are available in the United States [see Table 12]. Although more
potent on a milligram-per-milligram basis, the second-generation agents are not more
efficacious than the first-generation drugs at maximal doses. As many as 20 percent
of NIDDM patients who begin sulfonylurea therapy do not have an adequate
hypoglycemic response to maximal doses (these patients are termed primary
failures), and each year, an additional five to 10 percent of patients who initially
responded stop responding to sulfonylurea therapy (secondary failures). Many of the
drug interactions and side effects of the first-generation sulfonylureas do not occur or
are reduced with the second-generation agents. Hypoglycemia secondary to
sulfonylureas is infrequent; however, it is often more prolonged and more dangerous
than hypoglycemia secondary to insulin. An increased frequency of hypoglycemia
has been noted in patients receiving initial doses of the second-generation
sulfonylureas, and care should be taken to start patients on low doses, which can be
increased as needed. Weight gain often accompanies treatment with sulfonylureas.
Finally, the multicenter University Group Diabetes Program documented an
increased risk of cardiovascular mortality associated with tolbutamide,116 an effect
that is widely assumed to be a class action of these drugs.
The biguanide metformin is a relatively new addition to the medications available in
the United States for the treatment of NIDDM, although it has been widely used in
Canada and Europe for more than 20 years.117 It has a different mechanism of action
from that of sulfonylureas-it predominantly decreases hepatic glucose output. Unlike
the sulfonylureas, metformin is effective in both obese and nonobese patients and is
not associated with hypoglycemia or weight gain. Its potency is similar to that of the
sulfonylureas, lowering HbA1c by one to two percent. The most common adverse
effect of metformin is gastrointestinal; it causes abdominal distention, discomfort, or
even diarrhea in approximately 10 percent of patients. This side effect can be
decreased by advancing the dose slowly. Lactic acidosis, a potentially fatal side
effect that occurred more commonly with the biguanide phenformin, rarely occurs
with metformin (three episodes per 100,000 patient-years). The risk of lactic acidosis
is increased in patients with decreased glomerular filtration rate (serum creatinine >
1.4 mg/dl in women and > 1.5 mg/dl in men) and with circulatory failure (e.g.,
congestive heart failure or shock). Because of the potentially fatal consequences of
lactic acidosis, patients must be carefully selected and followed. Metformin can be
used as monotherapy or in combination with sulfonylurea in patients in whom
sulfonylurea therapy failed.118
Another relatively new oral agent, which can be used to treat IDDM and NIDDM, is
the -glycosidase inhibitor acarbose. This nonabsorbed drug acts like dietary fiber,
decreasing the rate of absorption of carbohydrate by competitively inhibiting the
intestinal wall enzymes that digest polysaccharides into absorbable monosaccharides.
The result is a blunted glycemic excursion in the postprandial period. Acarbose is
somewhat less potent than the sulfonylureas and metformin, lowering HbA1c levels
by 0.5 to 1.5 percent. However, it is effective in all diabetic patients, including
patients with IDDM, and may be used as monotherapy or in combination with any of
the other available hypoglycemic medications.119 When used as monotherapy,
acarbose is not associated with hypoglycemia or weight gain. The delivery of
undigested polysaccharides to the large intestine causes an increase in intestinal gas
formation, which decreases the drug's acceptability for some patients.
The newest therapy for NIDDM is the thiazolidinedione troglitazone, an
insulinomimetic agent. In dosages of 200 to 800 mg daily, troglitazone lowers HbA1c
by 0.5 to 1.0 percent and improves triglyceride levels with less insulin and without
significant weight gain or other serious side effects.120
Insulin is the most effective hypoglycemic agent, and because there is no upper limit
for insulin doses, an effective dose can almost always be found. In NIDDM, the total
dose of insulin is probably more important than the profile of its delivery, in contrast
to IDDM, where a physiologic profile is critical. Relatively high doses (50 to more
than 100 U/day) are generally required to achieve near-normal glycemia in obese
NIDDM patients.54,55, 121-123 Insulin can be administered as intermediate or long-acting
formulations in the morning or at bedtime, as so-called mixed-split regimens
(intermediate and rapid-acting) before breakfast or before breakfast and dinner, or as
fixed-combination formulations. Although some patients with NIDDM, such as those
who are not overweight, those who have profound hyperglycemia with weight loss or
ketosis, or those with very high triglyceride levels, should be started on insulin as
first-line therapy, most physicians reserve insulin for when the relatively weaker oral
agents have failed. Because the majority of NIDDM patients who start on oral agents
will eventually require insulin to maintain near-normal glycemia, the oral agents may
be viewed as temporizing measures.
The resistance to using insulin more frequently or more aggressively, by physicians
and patients alike, presumably resides in the need to inject it and fear of its side
effects, including hypoglycemia and weight gain. In fact, even when used in high
doses, insulin therapy rarely causes severe hypoglycemia in NIDDM patients.122,123
Moreover, most studies comparing insulin injection and oral agent sulfonylurea
therapy in NIDDM demonstrate similar levels of compliance.122, 124 Obviously,
patient preferences must be taken into account and the choice of therapeutic agents
must be individualized.
Combination therapy has been proposed for treating NIDDM, usually when
monotherapy has failed. One proposed strategy combines sulfonylureas and insulin.
Using the in sulin-sparing effect of sulfonylureas, this treatment attains similar levels
of glycemia with insulin doses that are five to 20 percent lower than the dose of
insulin required if used alone.125 Although the same metabolic results can be
achieved with higher insulin doses, some investigators have hypothesized that higher
insulin levels (or doses) may be atherogenic and should be avoided. The association
between insulin level and coronary artery disease was established in several large
population studies that excluded patients with diabetes, and there are few data that
suggest lowering insulin levels affects NIDDM outcomes positively. Furthermore,
although insulin doses can be lowered with combination therapy, insulin levels after
combination therapy are not significantly different from insulin levels after
monotherapy with insulin. Thus, given the added expense and the potential for
increased drug interactions, side effects, and complications and for decreased
compliance when two drugs are used, combination therapy with insulin and
sulfonylureas cannot be recommended routinely.
Other combinations, including metformin and sulfonylureas118 or acarbose and any of
the other hypoglycemic agents,119 can be used effectively, albeit at greater expense
and increased risk of hypoglycemia than with insulin alone.
EXERCISE
Exercise prescriptions for NIDDM are intended to aid weight loss, increase insulin
sensitivity, and promote cardiovascular conditioning. Because patients with NIDDM
are at increased risk for cardiovascular disease, patients who are beginning a new
exercise regimen should be given electrocardiograms and, if necessary, exercise
stress tests. Like IDDM patients with peripheral neuropathy, NIDDM patients with
neuropathic, insensate feet and limited circulation are at risk for foot trauma during
high-impact sports.
INNOVATIVE AND EXPERIMENTAL THERAPY FOR NIDDM
Several new therapeutic principles to treat NIDDM have been developed. Glucagon-
like peptide-I-(7-37) (GLP-I-[7-37]), a naturally occurring polypeptide that is
processed by the intestinal L cells from the preproglucagon molecule, has been
demonstrated to have potent effects on insulin secretion and synthesis. In addition,
GLP-I-(7-37) inhibits glucagon secretion and probably slows gastric emptying,
further contributing to its glucose-lowering effect. Intravenous or subcutaneous
administration has been shown to improve hyperglycemia in NIDDM without the
risk of hypoglycemia.126 Finally, a large multicenter study, the Diabetes Prevention
Program, has been initiated to determine whether NIDDM can be prevented or
delayed in volunteers who are at high risk by virtue of having impaired glucose
tolerance.
HYPEROSMOLAR, HYPERGLYCEMIC NONKETOTIC COMA
Hyperosmolar, hyperglycemic nonketotic coma is associated with severe
hyperglycemia (> 600 mg/dl) but, by definition, is not associated with ketosis or
ketoacidosis. Volume depletion is severe (the average fluid deficit is 25 percent of
total body water), and the serum osmolality is greater than 350 mOsm/kg. Usually,
either an illness or a drug or nutritional therapy (e.g., steroid treatment or tube
feeding with concentrated carbohydrate solutions) acutely exacerbates
hyperglycemia; the failure or inability to replace fluid loss precipitates profound
dehydration, obtundation, and, finally, coma. Precipitants, including infections, must
be identified and treated. The hyperosmolar dehydration is treated aggressively with
hypotonic (one-half normal) saline to replete both solute and free water. A low-dose
insulin infusion is effective in lowering glucose levels. In contrast to diabetic
ketoacidosis, mortality from hyperosmolar, hyperglycemic nonketotic coma may be
as high as 50 percent, perhaps reflecting the age of those affected and the severity of
the precipitating illness. After the initial episode is treated, patients often require no
insulin.
Complications
All forms of diabetes mellitus are accompanied by a similar spectrum of
microvascular and neurologic complications. Complications such as retinopathy and
nephropathy are specific for diabetes; in contrast, macrovascular complications,
including cardiac, peripheral, and cerebrovascular disease, also occur in the
nondiabetic population. The risk for developing macrovascular diseases in the
diabetic population, however, is much greater than the underlying risk in the general
population.
The development of long-term complications in the diabetic patient changes a
relatively benign, albeit difficult-to-manage, metabolic condition into a devastating
disease with profound morbidity and mortality. The clinical course for the IDDM
population is likely to include visual impairment or blindness, renal failure, and
vascular disorders [see Table 13]. The risk of experiencing these complications is
reduced 35 to 76 percent with intensive diabetes management [see Glycemic Goals
of Therapy: The DCCT, above].
Although the incidence of specific complications may differ somewhat from one
form of diabetes to another, the clinical expression of these complications in the
different clinical forms is identical. The cause of the myriad complications is
unknown. However, because they occur in forms of diabetes that are genetically
distinct but have disordered metabolism in common, some metabolic feature of
diabetes most likely causes the complications. Whether hyperglycemia per se causes
all of the complications, mediated perhaps through glycosylation of membrane
proteins or through other mechanisms, is unknown. The central role of
hyperglycemia in causing diabetes-specific complications (the glucose hypothesis) is
supported by interventional studies in animal models of diabetes and, most
convincingly, by the results of the DCCT. A DCCT-like study in Japanese patients
supports a similar benefit of intensive therapy on long-term complications in
NIDDM.108
Risk factors other than hyperglycemia have been identified. The most potent risk
factor is duration of diabetes. Hypertension is a risk factor for nephropathy and
probably retinopathy; pregnancy (in IDDM) is a risk factor for progression of
retinopathy.127,128 Finally, there may be a genetic component to the risk of developing
long-term complications.
RETINOPATHY
Given a long enough duration of diabetes, retinopathy occurred in almost all IDDM
and most NIDDM patients in the pre-DCCT era. After 15 years' duration, with
conventional diabetes management, more than 90 percent of IDDM patients had
clinical retinopathy.127 The most common form of retinopathy is nonproliferative
(also termed background retinopathy). It begins with loss of pericytes, the supporting
cells of the retinal vasculature, and progresses to characteristic microaneurysms that
are visible by direct ophthalmoscopy [see Figure 9a]. The microaneurysms, which
measure 50 to 100 æm in diameter, can occur anywhere in the retina but tend to
cluster near the macula, the part of the retina responsible for central vision and visual
acuity. Small punctate (or so-called dot-and-blot) hemorrhages form when the
microaneurysms leak blood. Leakage of serum leads to the formation of hard
exudates. These lesions are generally benign unless they occur near the macula and
in sufficient number to cause macular edema, which can decrease central vision and
acuity.

Figure 9a

Nonproliferative
Retinopathy
As retinopathy progresses, terminal capillaries become obstructed, and the retina
becomes ischemic. Infarctions of the nerve layer appear as soft (so-called cotton-
wool) exudates. In response to ischemia, new vessels proliferate from the surface of
the retina into the vitreous, giving the disorder its name, proliferative retinopathy
[see Figure 9b]. The new vessels develop in response to a specific angiogenic
growth factor released by the ischemic retina.129 They are extremely fragile and can
bleed into the vitreous, causing temporary loss of vision until the blood is
reabsorbed. Proliferative vessels that are greater than one quarter of the disk diameter
in size and that occur within one disk diameter of the disk are especially likely to
bleed. When the vitreous blood is reabsorbed, scars form, and traction on the retina
can lead to retinal detachment with profound and often permanent loss of vision.
 Figure 9b

Proliferative
Retinopathy
Laser treatment of proliferative retinopathy and macular edema has been
demonstrated to preserve vision.130,131 The role of the internist and ophthalmologist is
to detect and treat retinopathy requiring laser therapy before irreversible damage
occurs. Although fundus photography is the most sensitive means of detecting
retinopathy, ophthalmologists and even well-trained internists can detect
retinopathy.132 An examination with the pupil dilated is preferable. Because
retinopathy virtually never occurs before five years' duration in IDDM, regular
ophthalmologic examinations need not be performed until five years after diabetes
onset. Because the duration of NIDDM is much more difficult to assess accurately,
annual examinations are recommended, commencing at the time of diabetes
diagnosis. Pregnancy is a recognized risk factor for progression of retinopathy in
IDDM, and frequent eye examinations are recommended during pregnancy. Finally,
for patients with significant vitreous scarring and debris and retinal detachments,
vitrectomy can be performed. In this high-risk procedure, fibroproliferative scars are
excised, the retina is reattached if possible, and the debris-filled vitreous is replaced
with a salt solution. Vitrectomy can restore vision in selected cases.133
NEPHROPATHY
Nephropathy is the diabetic complication associated with the highest mortality.
Whereas the vast majority of diabetic patients developed some degree of retinopathy
in the pre-DCCT era, only 35 to 45 percent of IDDM patients and fewer than 20
percent of NIDDM patients developed clinical nephropathy.134,135 The progression of
nephropathy has been identified. It begins with low levels of albumin excretion, or
microalbuminuria (urine albumin levels of 30 to 300 mg/24 hr), as early as five years
after IDDM onset. This so-called incipient nephropathy progresses to
macroalbuminuria (i.e, urine albumin levels > 300 mg/24 hr as measured by a
positive dipstick proteinuria test or with a 24-hour collection) with accompanying
hypertension after IDDM of 12 to 20 years' duration. Ultimately, the nephrotic
syndrome and the decrease in glomerular filtration rate (GFR) progress to end-stage
renal disease (ESRD) [see Figure 10].136,137 Unlike the risk of retinopathy, the risk of
nephropathy does not continue to rise with increasing duration.127 If dipstick
proteinuria, the most reliable indicator of diabetic nephropathy, has not occurred
after 25 to 30 years of diabetes, the risk of developing nephropathy begins to
decrease. Duration of diabetes and, perhaps, family history of hypertension are risk
factors for nephropathy,138 whereas glucose control has not been clearly identified as
a risk factor. Controlling hypertension and using low-protein diets can decrease the
rate of GFR decline late in the course of nephropathy.139-141 Angiotensin-converting
enzyme inhibitors have been demonstrated to play an especially important role in
reducing progression from microalbuminuria and from more advanced stages of
nephropathy in IDDM and NIDDM.142-144 Intensive diabetes treatment slows and
perhaps prevents the development of microalbuminuria and clinical-grade, dipstick-
positive proteinuria (urine albumin levels > 300 mg/24 hr).8, 145,146 Whether intensive
diabetes treatment affects the ultimate progression of microalbuminuria to ESRD is
unknown.

Figure 10

Progression of
Nephropathy in
IDDM
Diabetic patients with ESRD can be treated with transplantation, hemodialysis, or
peritoneal dialysis. Hemodialysis is associated with an increased risk of
cardiovascular events and with a risk of retinal bleeding caused by profound blood
pressure fluctuations and anticoagulation. Thus, transplantation and peritoneal
dialysis are the preferred treatments.
NEUROPATHY
Neuropathy has protean manifestations in diabetes. The most common manifestation
is the peripheral, symmetric sensorimotor neuropathy that presents as numbness or
tingling in the toes or feet. The symptoms are usually only mildly disturbing and
require no specific treatment. Patients often report that these symptoms abate over
time, as the neuropathy becomes more severe and hypoesthesia or anesthesia takes
the place of paresthesias and dysesthesias. The insensate feet become vulnerable to
trauma, and neuropathic foot ulcers are all too frequent causes of hospitalization and
amputation. In rare instances, peripheral neuropathies can be painful, with
dysesthesias interfering with sleep and activity. Tricyclic antidepressants, phenytoin,
and carbamazepine are all somewhat effective. Capsaicin, a topical treatment that
depletes substance P, a pain-modulating neurotransmitter, from type C nociceptive
fibers, has been demonstrated to have some efficacy.147 Narcotics should be avoided
because the potential for addiction is high. The DCCT demonstrated a major
beneficial impact of intensive therapy on diabetic neuropathy.8 For patients whose
feet are at risk for ulceration, foot care education and inspection have been
demonstrated to reduce foot trauma and hospitalizations.148
Other forms of diabetic neuropathy include mononeuropathies, entrapment
syndromes, and autonomic neuropathy. The mononeuropathies, which are generally
asymmetric, affect cranial or peripheral nerves and are thought to be secondary to
vascular occlusion leading to nerve infarcts. Entrapment syndromes, such as carpal
tunnel syndrome, occur more frequently in diabetics than in nondiabetics. Finally,
autonomic neuropathy can affect the gastrointestinal, cardiovascular, or
genitourinary systems. The clinical syndromes include diabetic gastroparesis
characterized by delayed gastric emptying: early satiety and fullness and nausea and
vomiting. Diabetic diarrhea results in nocturnal diarrhea and incontinence alternating
with constipation. Cardiovascular neuropathy produces resting tachycardia with
postural hypotension, and genitourinary neuropathy causes impotence and hypotonia
or atonia of the bladder resulting in overflow incontinence. Gastroparesis is
effectively treated with metoclopramide and diabetic diarrhea with clonidine149,150;
impotence can be treated with penile injections or intraurethral administration of
vasodilators or with implantable prostheses or vacuum devices.151,152
CARDIOVASCULAR DISEASE
Cardiovascular disease is not specific for diabetes. However, the risk of cardiac
disease is increased two to almost 10-fold in NIDDM.153 The relative risk of
cardiovascular disease for diabetic women (compared with nondiabetic women) is
increased even more than for diabetic men, essentially erasing the protective effect
usually conferred by estrogen. Numerous risk factors for cardiovascular disease
accompany NIDDM, including obesity, hypertension, dyslipidemia (low HDLs and
high VLDLs with small, dense LDLs), and hyperglycemia. A study of survivors in
the original Framingham Study cohort has documented that the HbA1c level is
associated with the occurrence of cardiac disease in women, independent of other
recognized risk factors.154 Development of nephropathy in IDDM or NIDDM confers
an especially high risk of coronary artery disease.
The incidence of peripheral vascular and cerebrovascular disease is also increased in
patients with diabetes, and these disorders are a source of significant morbidity and
mortality among diabetic patients. The combination of peripheral neuropathy and
peripheral vascular disease results in a risk of amputations in the diabetic population
that is 40 times greater than that in the nondiabetic population.
DAVID M. NATHAN, M.D.
References

1.Banting FG, Best CH, Collip JB, et al: Pancreatic extracts in the treatment of diabetes mellitus. Can Med Assoc J
12:141, 1922
2.Burrow GN, Hazlett BE, Phillips MJ: A case of diabetes mellitus. N Engl J Med 306:340, 1982 [PMID
7033790]

3.Zinman B: The physiologic replacement of insulin. N Engl J Med 321:363, 1989

4.Howey DC, Bowsher RR, Brunelle RL, et al: [Lys(B28), Pro(B29)]-Human insulin. Diabetes 43:396, 1994
[PMID 8314011]

5.Nathan DM, Fogel H, Norman D, et al: Long-term metabolic and quality of life results with pancreatic/renal
transplantation in insulin-dependent diabetes mellitus. Transplantation 52:85, 1991 [PMID 1858159]

6.Robertson AP: Pancreatic and islet transplantation for diabetes-cures or curiosities? N Engl J Med 327:1861,
1992

7.Lanza RP, Sullivan SJ, Chick WL: Islet transplantation with immunoisolation. Diabetes 41:1503, 1992 [PMID
1446791]

8.The effect of intensive treatment of diabetes on the development and progression of long-term complications in
insulin-dependent diabetes mellitus. Diabetes Control and Complications Trial Research Group. N Engl J
Med 329:977, 1993

9.Gough DA, Armour JC: Development of the implant glucose sensor-what are the prospects and why is it taking
so long? Diabetes 44:1005, 1995

10. Selam J-L, Micossi P, Dunn FL, et al: Clinical trial of programmable implantable insulin pump for type I
diabetes. Diabetes Care 15:877, 1992

11. Himsworth HP: Diabetes mellitus: its differentiation into insulin-sensitive and insulin insensitive types.
Lancet 1:117, 1936
12. Eisenbarth GS: Type I diabetes mellitus: a chronic autoimmune disease. N Engl J Med 314:1360, 1986

13. Leahy JL: Natural history of -cell dysfunction in NIDDM. Diabetes Care 13:992, 1990

14. National Diabetes Data Group: Classification and diagnosis of diabetes mellitus and other categories of
glucose intolerance. Diabetes 28:1039, 1979

15. Neufield M, Maclaren NK, Blizzard RM: Two types of autoimmune Addison's disease associated with
different polyglandular autoimmune (PGA) syndromes. Medicine (Baltimore) 60:355, 1981

16. Report of the Expert Committee on the diagnosis and classification of diabetes mellitus. Expert Committee on
the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 20:1183, 1997

17. LaPorte R, Matsushima M, Chang YF: Prevalence and incidence of insulin-dependent diabetes. Diabetes in
America, 2nd ed. Harris M, Ed. National Institutes of Health, Bethesda, Maryland, NIH Publication No 95-
1468, 1995, p 37 [see Web link]

18. Harris MI, Hadden WC, Knowler WC, et al: Prevalence of diabetes and impaired glucose tolerance and
plasma glucose levels in U.S. population aged 20-74 yr. Diabetes 36:523, 1987

19. King H, Zimmet P: Trends in the incidence and prevalence of diabetes: non-insulin dependent diabetes
mellitus. World Health Stat Q 41:190, 1988 [PMID 2466380]

20. Bennett PH, Bogardus C, Tuomilehto J, et al: Epidemiology and natural history of NIDDM: nonobese and
obese. International Textbook of Diabetes. Alberti KGMM, DeFronzo RA, Keen H, et al, Eds. John Wiley &
Sons, Ltd, Chichester, England, 1992, p 148

21. Hartz AJ, Rupley DC, Rimm AA: The association of girth measurements with disease in 32,856 women. Am
J Epidemiol 119:71, 1984 [PMID 6691337]

22. Savage PJ, Bennett PH, Senter RG, et al: High prevalence of diabetes in young Pima Indians: evidence of
phenotypic variation in a genetically isolated population. Diabetes 28:937, 1979 [PMID 478185]

23. Zimmet P, Taft P, Guinea A, et al: The high prevalence of diabetes mellitus on a Central Pacific island.
Diabetologia 13:111, 1977 [PMID 852640]

24. Rowley MJ, MacKay IR, Chen Q-Y, et al: Antibodies to glutamic acid decarboxylase discriminate major
types of diabetes mellitus. Diabetes 41:548, 1992 [PMID 1607079]

25. Greenbaum CJ, Palmer JP, Nagataki S, et al: Participating labs improved specificity of ICA assays in the
Fourth International Immunology of Diabetes Exchange Workshop. Diabetes 41:1570, 1992 [PMID
1446798]

26. Davies JL, Kawaguchi Y, Bennett ST, et al: A genome-wide search for human type 1 diabetes susceptibility
genes. Nature 371:130, 1994 [PMID 8072542]

27. Oats JN, Beischer NA, Grant PT: The emergence of diabetes and impaired glucose tolerance in women who
had gestational diabetes. Gestational Diabetes. Weiss PA, Coustan DR, Eds. Springer-Verlag, New York,
1988, p 199

28. Tattersall RB: Mild familial diabetes with dominant inheritance. Q J Med 43:339, 1974

29. Froguel P, Vaxillaire M, Sun F, et al: Close linkage of glucokinase locus on chromosome 7p to early-onset
non-insulin-dependent diabetes mellitus. Nature 356:162, 1992 [PMID 1545870]

30. Kadowaki T, Kadowaki H, Mori Y, et al: A subtype of diabetes mellitus associated with a mutation of
mitochondrial DNA. N Engl J Med 330:962, 1994
31. Jarrett JJ, McCartney P, Keen H: The Bedford survey: ten year mortality rates in newly diagnosed diabetics,
borderline diabetics and normoglycaemic controls and risk indices for coronary heart disease in borderline
diabetics. Diabetologia 22:79, 1982

32. Fuller JH, McCartney P, Jarrett RJ, et al: Hyperglycaemia and coronary heart disease: the Whitehall Study. J
Chron Dis 32:721, 1979

33. Ziegler AG, Herskowitz RD, Jackson RA, et al: Predicting type I diabetes. Diabetes Care 13:762, 1990
[PMID 2201499]

34. Faustman D, Eisenbarth G, Daley J, et al: Abnormal T-lymphocyte subsets in type I diabetes. Diabetes
38:1462, 1989 [PMID 2533573]

35. Palmer JP, McCulloch DK: Prediction and prevention of IDDM-1991. Diabetes 40:943, 1991 [PMID
1860558]

36. Baekkeskov S, Aanstoot H-J, Christgau S, et al: Identification of the 64K autoantigen in insulin-dependent
diabetes as the GABA-synthesizing enzyme glutamic acid decarboxylase. Nature 347:151, 1990 [PMID
1697648]

37. Verge CF, Gianani R, Kawasaki E, et al: Prediction of type I diabetes in first-degree relatives using a
combination of insulin, GAD, and ICA512bdc/IA-2 autoantibodies. Diabetes 45:926, 1996 [PMID 8666144]

38. Roep BO: T-cell responses to autoantigens in IDDM: the search for the Holy Grail. Diabetes 45:1147, 1996

39. Barnett AH, Eff C, Leslie RDG, et al: Diabetes in identical twins: a study of 200 pairs. Diabetologia 20:87,
1981 [PMID 7193616]

40. Thomson G, Robinson WP, Kuhner MK: Genetic heterogeneity, modes of inheritance, and risk estimates for
a joint study of Caucasians with insulin-dependent diabetes mellitus. Am J Hum Genet 43:799, 1988 [PMID
3057885]

41. Baisch JM, Weeks T, Giles R, et al: Analysis of HLA-DQ genotypes and susceptibility in insulin-dependent
diabetes mellitus. N Engl J Med 322:1836, 1990 [PMID 2348836]

42. Yalow RS, Berson SA: Plasma insulin concentrations in nondiabetic and early diabetic subjects:
determinations by a new sensitive immuno-assay technique. Diabetes 9:254, 1960

43. Yalow RS, Berson SA: Immunoassay of endogenous plasma insulin in man. J Clin Invest 39:1157, 1960

44. Rahier J, Goebbels RM, Henquin JC: Cellular composition of the human diabetic pancreas. Diabetologia
24:366, 1983 [PMID 6347784]

45. Reaven GM, Bernstein R, Davis B, et al: Nonketotic diabetes mellitus: insulin deficiency or insulin
resistance? Am J Med 60:80, 1976

46. Moller DE, Flier JS: Insulin resistance-mechanisms, syndromes, and implications. N Engl J Med 325:938,
1991 [PMID 1881419]

47. Sun XJ, Rothenberg P, Kahn CR, et al: Structure of the insulin receptor substrate IRS-1 defines a unique
signal transduction protein. Nature 352:73, 1991 [PMID 1648180]

48. DeFronzo RA: The triumvirate: -cell, muscle, liver: a collusion responsible for NIDDM. Diabetes 37:667,
1988

49. Brunzell JD, Robertson RP, Lerner RL, et al: Relationships between fasting plasma glucose levels and insulin
secretion during intravenous glucose tolerance tests. J Clin Endocrinol Metab 42:222, 1976 [PMID 1262429]
50. Yoshioka N, Kuzuya T, Matsuda A, et al: Effects of dietary treatment on serum insulin and proinsulin
response in newly diagnosed NIDDM. Diabetes 38:262, 1989 [PMID 2644145]

51. Cerasi E, Luft R: The plasma insulin response to glucose infusion in healthy subjects and in diabetes mellitus.
Acta Endocrinol 55:278, 1967 [PMID 5338206]

52. Leahy JL, Bonner-Weir S, Weir GC: Abnormal glucose regulation of insulin secretion in models of reduced
B-cell mass. Diabetes 33:667, 1984 [PMID 6203798]

53. Turner RC, Holman RR: Beta cell function during insulin or chlorpropamide treatment of maturity-onset
diabetes mellitus. Diabetes 27(suppl 1):241, 1978

54. Andrews WJ, Vasquez B, Nagulesparan M, et al: Insulin therapy in obese, non-insulin-dependent diabetes
induces improvements in insulin action and secretion that are maintained for two weeks after insulin
withdrawal. Diabetes 33:634, 1984 [PMID 6376219]

55. Garvey WT, Olefsky JM, Griffin J, et al: The effect of insulin treatment on insulin secretion and insulin
action in type II diabetes mellitus. Diabetes 34:222, 1985 [PMID 3882489]

56. Blackshear PJ, Shulman GI, Roussell AM, et al: Metabolic response to three years of continuous, basal rate
intravenous insulin infusion in type II diabetic patients. J Clin Endocrinol Metab 61:753, 1985 [PMID
3897260]

57. Lillioja S, Mott DM, Howard BV, et al: Impaired glucose tolerance as a disorder of insulin action:
longitudinal and cross-sectional studies in Pima Indians. N Engl J Med 318:1217, 1988 [PMID 3283552]

58. Warram JH, Martin BC, Krolewski AS, et al: Slow glucose removal rate and hyperinsulinemia precede the
development of type II diabetes in the offspring of diabetic parents. Ann Intern Med 113:909, 1990 [PMID
2240915]

59. O'Rahilly S, Turner RC, Matthews DR: Impaired pulsatile secretion of insulin in relatives of patients with
non-insulin-dependent diabetes. N Engl J Med 318:1225, 1988 [PMID 3283553]

60. Eriksson J, Franssila-Kallunki A, Ekstrand A, et al: Early metabolic defects in persons at increased risk for
non-insulin-dependent diabetes mellitus. N Engl J Med 321:337, 1989 [PMID 2664520]

61. Pimenta W, Korytkowski M, Mitrakou A, et al: Pancreatic beta-cell dysfunction as the primary genetic lesion
in NIDDM: evidence from studies in normal glucose-tolerant individuals with a first-degree NIDDM relative.
JAMA 273:1855, 1995 [PMID 7776502]

62. Ullrich A, Bell JR, Chen EY, et al: Human insulin receptor and its relationship to the tyrosine kinase family
of oncogenes. Nature 313:756, 1985 [PMID 2983222]

63. Foster DW, McGarry JD: The metabolic derangements and treatment of diabetic ketoacidosis. N Engl J Med
309:159, 1983 [PMID 6408476]

64. Gerich JE, Lorenzi M, Bier DM, et al: Prevention of human diabetic ketoacidosis by somatostatin: evidence
for an essential role of glucagon. N Engl J Med 292:985, 1975 [PMID 804137]

65. Effects of age, duration and treatment of insulin-dependent diabetes mellitus on residual -cell function:
observations during eligibility testing for the Diabetes Control and Complications Trial (DCCT). The DCCT
Research Group. J Clin Endocrinol Metab 65:30, 1987

66. Podolsky S: Hyperosmolar nonketotic coma in the elderly diabetic. Med Clin North Am 62:815, 1978

67. Mitrakou A, Kelley D, Veneman T, et al: Contribution of abnormal muscle and liver glucose metabolism to
postprandial hyperglycemia in NIDDM. Diabetes 39:1381, 1990 [PMID 2121568]
68. Holman RR, Turner RC: The basal plasma glucose: a simple relevant index of maturity-onset diabetes. Clin
Endocrinol 14:279, 1980

69. Nathan DM, Singer DE, Godine JE, et al: Retinopathy in older type II diabetics: association with glucose
control. Diabetes 35:797, 1986 [PMID 3721064]

70. Walden CE, Knopp RH, Wahl PW, et al: Sex differences in the effect of diabetes mellitus on lipoprotein
triglyceride and cholesterol concentrations. N Engl J Med 311:953, 1984 [PMID 6472421]

71. Singer DE, Samet JH, Coley CM, et al: Screening for diabetes mellitus. Ann Intern Med 109:639, 1988
[PMID 3138934]

72. Cahill GF Jr, Etzwiler LD, Freinkel N: "Control" and diabetes (editorial). N Engl J Med 294:1004, 1976
[PMID 1256496]

73. Siperstein MD, Foster DW, Knowles HC, et al: Control of blood glucose and diabetic vascular disease
(editorial). N Engl J Med 296:1060, 1977

74. Ingelfinger FJ: Debates on diabetes. N Engl J Med 296:1228, 1977

75. Ballantyne AJ, Loewenstein A: The pathology of diabetic retinopathy. Trans Ophthalmol UK 63:95, 1943

76. Kimmelstiel P, Wilson C: Intercapillary lesions in the glomeruli of the kidney. Am J Pathol 12:83, 1936

77. Engerman R, Bloodworth JM Jr, Nelson S: Relationship of microvascular disease in diabetes to metabolic
control. Diabetes 26:760, 1977

78. Klein R, Klein BE, Moss SE, et al: The Wisconsin epidemiologic study of diabetic retinopathy: II. Prevalence
and risk of diabetic retinopathy when age at diagnosis is less than 30 years. Arch Ophthalmol 102:520, 1984

79. Nathan DM: Modern management of insulin-dependent diabetes mellitus. Med Clin North Am 72:1365, 1988

80. Nathan DM, Singer DE, Hurxthal K, et al: The clinical information value of the glycosylated hemoglobin
assay. N Engl J Med 310:341, 1984

81. Management of diabetes mellitus in pregnancy. ACOG Technical Bulletin 92:1, 1986

82. Fuhrmann K, Reiher H, Semmler K, et al: Prevention of congenital malformations in infants of insulin-
dependent diabetic mothers. Diabetes Care 6:219, 1983

83. Jovanovic L, Druzin M, Peterson CM: Effect of euglycemia on the outcome of pregnancy in insulin-
dependent diabetic women as compared with normal control subjects. Am J Med 71:921, 1981 [PMID
7032287]

84. Nathan DM: Inferences and implications: do results from the Diabetes Control and Complications Trial apply
in NIDDM? Diabetes Care 18:251, 1995

85. Malone JI, Rosenbloom AL, Grgic A, et al: The role of urine sugar in diabetic management. Am J Dis Child
130:1324, 1976 [PMID 998574]

86. Frishman D, Ardito DM, Graham SM: Performance of glucose monitors. Lab Med 23:179, 1992

87. Larsen ML, Horder M, Mogensen EF: Effect of long-term monitoring of glycosylated hemoglobin levels in
insulin-dependent diabetes mellitus. N Engl J Med 323:1021, 1990 [PMID 2215560]

88. Halle J-P, Lambert J, Lindmayer I, et al: Twice-daily mixed regular and NPH insulin injections with new jet
injector versus conventional syringes: pharmacokinetics of insulin absorption. Diabetes Care 9:279, 1986
89. Rizza RA, Gerich JE, Haymond MW, et al: Control of blood sugar in insulin-dependent diabetes: comparison
of an artificial endocrine pancreas, continuous subcutaneous insulin infusion, and intensified conventional
insulin therapy. N Engl J Med 303:1313, 1980 [PMID 7001229]

90. Nathan DM, Lou P, Avruch J: Intensive conventional and insulin pump therapies in adult type I diabetes: a
crossover study. Ann Intern Med 97:31, 1982 [PMID 7046554]

91. Epidemiology of severe hypoglycemia in the Diabetes Control and Complications Trial. The DCCT Research
Group. Am J Med 90:450, 1991

92. Nathan DM, Dunn FL, Bruch J, et al: Postprandial insulin profiles with implantable pump therapy may
explain decreased frequency of severe hypoglycemia, compared with intensive subcutaneous regimens, in
insulin-dependent diabetes mellitus patients. Am J Med 100:412, 1996 [PMID 8610727]

93. Nutritional recommendations and principles for individuals with diabetes mellitus: 1986 position statement.
American Diabetes Association. Diabetes Care 10:126, 1987

94. Jenkins DJA, Wolever TMS, Wong GS, et al: Glycemic responses to foods: possible differences between
insulin-dependent and noninsulin-dependent diabetics. Am J Clin Nutr 40:971, 1984

95. Faich GA, Fishbein HA, Ellis SE: The epidemiology of diabetic acidosis: a population-based study. Am J
Epidemiol 117:551, 1983 [PMID 6405612]

96. Page M, Alberti KGMM, Greenwood R, et al: Treatment of diabetic coma with continuous low-dose infusion
of insulin. Br Med J 2:687, 1974 [PMID 4855253]

97. Adrogue HJ, Wilson H, Boyd AE, et al: Plasma acid-base patterns in diabetic ketoacidosis. N Engl J Med
307:1603, 1982 [PMID 6815530]

98. Morris LR, Murphy MB, Kitabchi AE: Bicarbonate therapy in severe diabetic ketoacidosis. Ann Intern Med
105:836, 1986 [PMID 3096181]

99. Keller U, Berger W: Prevention of hypophosphatemia by phosphate infusion during treatment of diabetic
ketoacidosis and hyperosmolar coma. Diabetes 29:87, 1980 [PMID 6766411]

100. Cryer PE, Gerich JE: Glucose counterregulation, hypoglycemia, and intensive insulin therapy in diabetes
mellitus. N Engl J Med 313:232, 1985 [PMID 2861565]

101. Consensus statement on self-monitoring of blood glucose. American Diabetes Association. Diabetes Care
10:95, 1987

102. Bolli G, De Feo P, Compagnucci P, et al: Abnormal glucose counterregulation in insulin-dependent diabetes
mellitus: interaction of anti-insulin antibodies and impaired glucagon and epinephrine secretion. Diabetes
32:134, 1983 [PMID 6337896]

103. Amiel SA, Tamborlane WV, Simonson DC, et al: Defective glucose counterregulation after strict glycemic
control of insulin-dependent diabetes mellitus. N Engl J Med 316:1376, 1987 [PMID 3553949]

104. Cranston I, Lomas J, Maran A, et al: Restoration of hypoglycaemia awareness in patients with long-duration
insulin-dependent diabetes. Lancet 344:283, 1994 [PMID 7914259]

105. Faustman D, Coe C: Prevention of xenograft rejection by masking donor HLA class I antigens. Science
252:1700, 1991 [PMID 1710828]

106. Scharp DW, Lacy PE, Santiago JV, et al: Insulin independence after islet transplantation into type I diabetic
patient. Diabetes 39:515, 1990 [PMID 2108071]
107. Cyclosporin-induced remission of IDDM after early intervention. The Canadian-European Randomized
Control Trial Group. Diabetes 37:1514, 1988

108. Ohkubo Y, Kishikawa H, Araki E, et al: Intensive insulin therapy prevents the progression of diabetic
microvascular complications in Japanese patients with non-insulin-dependent diabetes mellitus: a randomized
prospective 6-year study. Diabetes Res Clin Pract 28:103, 1995 [PMID 7587918]

109. Hadden DR, Montgomery DAD, Skelly RJ, et al: Maturity onset diabetes mellitus: response to intensive
dietary management. Br Med J 3:276, 1975 [PMID 1148773]

110. Long SD, O'Brien K, MacDonald KG Jr, et al: Weight loss in severely obese subjects prevents the
progression of impaired glucose tolerance to type II diabetes: a longitudinal interventional study. Diabetes
Care 17:372, 1994 [PMID 8062602]

111. Wing RR, Jeffery RW: Outpatient treatments of obesity: a comparison of methodology and clinical results.
Int J Obes 3:261, 1979 [PMID 395116]

112. Wing RR, Marcus MD, Epstein LH, et al: Type II diabetic subjects lose less weight than their overweight
nondiabetic spouses. Diabetes Care 10:563, 1987

113. Allen BT, Delong ER, Feussner JR: Impact of glucose self-monitoring on non-insulin-treated patients with
type II diabetes mellitus. Diabetes Care 13:1044, 1990

114. Singer DE, Coley CM, Samet JH, et al: Tests of glycemia in diabetes mellitus: their use in establishing
diagnosis and in treatment. Ann Intern Med 110:125, 1989 [PMID 2642375]

115. Groop LC: Sulfonylureas in NIDDM. Diabetes Care 15:737, 1992

116. A study of the effects of hypoglycemic agents on vascular complications in patients with adult onset diabetes:
supplementary report on nonfatal events in patients treated with tolbutamide. University Group Diabetes
Program. Diabetes 25:1129, 1976

117. Bailey CJ: Biguanides and NIDDM. Diabetes Care 15:755, 1992

118. DeFronzo RA, Goodman AM: Efficacy of metformin in patients with non-insulin-dependent diabetes
mellitus. N Engl J Med 333:541, 1995 [PMID 7623902]

119. Chiasson JL, Josse RG, Hunt JA, et al: The efficacy of acarbose in the treatment of patients with non-insulin-
dependent diabetes mellitus: a multicenter controlled clinical trial. Ann Intern Med 121:928, 1994 [PMID
7734015]

120. Iwamoto Y, Kosaka K, Kuzuya T, et al: Effects of troglitazone, a new hypoglycemic agent in patients with
NIDDM poorly controlled by diet therapy. Diabetes Care 19:151, 1996 [PMID 8718436]

121. Glaser B, Leibovich G, Nesher R, et al: Improved beta-cell function after intensive insulin treatment in severe
non-insulin-dependent diabetes. Acta Endocrinol 118:365, 1988 [PMID 3293339]

122. Nathan DM, Roussell A, Godine JE: Glyburide or insulin for metabolic control in non-insulin-dependent
diabetes mellitus: a randomized double-blind study. Ann Intern Med 108:334, 1988 [PMID 3124685]

123. Colwell JA: The feasibility of intensive insulin management in non-insulin-dependent diabetes mellitus:
implications of the Veterans Affairs Cooperative Study on Glycemic Control and Complications in NIDDM.
Ann Intern Med 124:131, 1996

124. Diehl AK, Sugarek NJ, Bauer RL: Medication compliance in non-insulin-dependent diabetes: a randomized
comparison of chlorpropamide and insulin. Diabetes Care 8:219, 1985 [PMID 3891264]
125. Peters AL, Davidson MB: Insulin plus a sulfonylurea agent for treating type 2 diabetes. Ann Intern Med
115:45, 1991 [PMID 1828656]

126. Nathan DM, Schreiber E, Fogel H, et al: Insulinotropic action of glucagon-like peptide-I-(7-37) in diabetic
and nondiabetic subjects. Diabetes Care 15:270, 1992 [PMID 1547685]

127. Krolewski AS, Warram JH, Rand LI, et al: Epidemiologic approach to the etiology of type I diabetes mellitus
and its complications. N Engl J Med 317:1390, 1987 [PMID 3317040]

128. Klein BEK, Moss SE, Klein R: Effect of pregnancy on progression of diabetic retinopathy. Diabetes Care
13:34, 1990

129. Palmberg P, Smith M, Waltman S, et al: The natural history of retinopathy in insulin-dependent juvenile-
onset diabetes. Ophthalmology 88:613, 1981 [PMID 7267029]

130. Aiello LP, Avery RL, Arrig PG, et al: Vascular endothelial growth factor in ocular fluid of patients with
diabetic retinopathy and other retinal disorders. N Engl J Med 331:1480, 1994 [PMID 7526212]

131. Photocoagulation treatment of proliferative diabetic retinopathy: clinical application of Diabetic Retinopathy
Study (DRS) findings, DRS report number 8. The Diabetic Retinopathy Study Research Group.
Ophthalmology 88:583, 1981

132. Early photocoagulation for diabetic retinopathy: ETDRS report number 9. Early Treatment Diabetic
Retinopathy Study Research Group. Ophthalmology 98:766, 1991

133. Nathan DM, Fogel HA, Godine JE, et al: Role of diabetologist in evaluating diabetic retinopathy. Diabetes
Care 14:26, 1991 [PMID 1991432]

134. Early vitrectomy for severe vitreous hemorrhage in diabetic retinopathy. Diabetic Retinopathy Vitrectomy
Study Group. Arch Ophthalmol 103:1644, 1985

135. Andersen AR, Christiansen JS, Andersen JK, et al: Diabetic nephropathy in type 1 (insulin-dependent)
diabetes: an epidemiological study. Diabetologia 25:496, 1982

136. Humphrey LL, Ballard DJ, Frohnert PP, et al: Chronic renal failure in non-insulin-dependent diabetes
mellitus: a population-based study in Rochester, Minnesota. Ann Intern Med 111:788, 1989 [PMID 2817626]

137. Viberti GC, Jarrett RJ, Mahmud U, et al: Microalbuminuria as a predictor of clinical nephropathy in insulin-
dependent diabetes mellitus. Lancet 1:1430, 1982 [PMID 6123720]

138. Krolewski AS, Canessa M, Warram JH, et al: Predisposition to hypertension and susceptibility to renal
disease in insulin-dependent diabetes mellitus. N Engl J Med 318:140, 1988 [PMID 3336401]

139. Cohen D, Dodds R, Viberti G: Effect of protein restriction in insulin-dependent diabetics at risk of
nephropathy. Br Med J 294:295, 1987

140. Mogensen CE: Long-term antihypertensive treatment inhibits progression of diabetic nephropathy. Br Med J
285:685, 1982

141. Vasquez B, Flock EV, Savage PJ, et al: Sustained reduction of proteinuria in type 2 (non-insulin-dependent)
diabetes following diet-induced reduction of hyperglycaemia. Diabetologia 26:127, 1984 [PMID 6714534]

142. Marre M, Chatellier G, Leblanc H, et al: Prevention of diabetic nephropathy with enalapril in normotensive
diabetics with microalbuminuria. BMJ 297:1092, 1988 [PMID 2848604]

143. Lewis EJ, Hunsicker LG, Bain RP, et al: The effect of angiotensin-converting-enzyme inhibition on diabetic
nephropathy. N Engl J Med 329:1456, 1993 [PMID 8413456]
 144. Sano T, Kawamura T, Matsumae H, et al: Effects of long-term enalapril treatment on persistent micro-
albuminuria in well-controlled hypertensive and normotensive NIDDM patients. Diabetes Care 17:420, 1994
[PMID 8062609]

145. Feldt-Rasmussen B, Mathiesen ER, Deckert T: Effect of two years of strict metabolic control on progression
of incipient nephropathy in insulin-dependent diabetes. Lancet 2:1300, 1986 [PMID 2878175]

146. Reichard P, Rosenqvist U: Nephropathy is delayed by intensified insulin treatment in patients with insulin-
dependent diabetes mellitus and retinopathy. J Intern Med 226:81, 1989 [PMID 2671247]

147. Donofrio PD, Walker FO, Hunt VP, et al: Treatment of painful diabetic neuropathy with topical capsaicin: a
multicenter, double-blind, vehicle controlled study. Arch Intern Med 151:2225, 1991

148. Kitzelman DK, Slemenda CW, Langefeld CD, et al: Reduction of lower extremity clinical abnormalities in
patients with non-insulin-dependent diabetes mellitus: a randomized, controlled trial. Ann Intern Med 119:36,
1993

149. Snape WJ, Battle WM, Schwartz SS, et al: Metoclopramide to treat gastroparesis due to diabetes mellitus.
Ann Intern Med 96:444, 1982 [PMID 7065559]

150. Fedorak RN, Field M, Chang EB: Treatment of diabetic diarrhea with clonidine. Ann Intern Med 102:197,
1985 [PMID 3966758]

151. Chen J, Godschalk M, Katz PG, et al: The lowest effective dose of prostaglandin E 1 as treatment for erectile
dysfunction. J Urol 153:80, 1995 [PMID 7966797]

152. Scott FB, Fishman IJ, Light JK: An inflatable penile prosthesis for treatment of diabetic impotence. Ann
Intern Med 92:340, 1980 [PMID 7356228]

153. Kannel WB, McGee DL: Diabetes and cardiovascular disease: the Framing ham Study. JAMA 241:2035,
1979 [PMID 430798]

154. Singer DE, Nathan DM, Anderson KM, et al: Association of HbA 1c with prevalent cardiovascular disease in
the original cohort of the Framingham Heart Study. Diabetes 41:202, 1992 [PMID 1733810]

Acknowledgments
Figures 1, 3, 6, 8, and 10 Janet Betries.
Figures 2, 4, 5, and 9 Andy Christie.
Table 1 Adapted from "Classification and Diagnosis of Diabetes Mellitus and Other Categories of Glucose Intolerance,"
by the National Diabetes Data Group, in Diabetes 28:1039, 1979. Used by permission.
Table 2 Data for impaired glucose tolerance derived from "Prevalence of Diabetes and Impaired Glucose Tolerance and
Plasma Glucose Levels in U.S. Population Aged 20-74 Yr," by M. I. Harris, W. C. Hadden, W. C. Knowler, et al, in
Diabetes 36:523, 1987. Used by permission.

Copyright © 1999, Scientific American, Inc. All Rights Reserved.

Figure 1. Stages of IDDM/Beta Cell Mass

The stages in the development of insulin-dependent diabetes mellitus (IDDM) are
plotted against the hypothetical beta cell mass. Although IDDM characteristically
has an acute clinical onset, that onset is preceded by a long, asymptomatic
preclinical period. This period may be triggered by some environmental event in
genetically predisposed individuals. During the preclinical period, insulin-producing
beta cells are progressively destroyed, and anti-islet cell antibodies (ICAs) appear as
long as a decade before the clinical appearance of diabetes. Clinical diabetes
appears when more than 90 percent of the islet beta cells have been destroyed or
when insulin requirements exceed the secretory capacity of the decreased beta cell
mass. Complete destruction of beta cells can be detected by the absence of C-peptide
from the serum. The age at onset and the time between events are variable; for
instance, overt diabetes can appear from five months to 10 years after the
appearance of immunologic abnormalities.
Copyright © 1999, Scientific American, Inc. All Rights Reserved.