| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
-Glucosidase Inhibition as Monotherapy in Elderly Type 2 Diabetic Patients
Bayer Pharmaceuticals (P.S.J., R.F.C., C.L.M.), West Haven, Connecticut 06516; SUNY Health Sciences Center, (H.E.L.), Brooklyn, New York 11203; Brigham and Women’s Hospital (D.C.S.), Boston, Massachusetts 02115; University of Texas Southwestern Medical Center (P.R.), Dallas, Texas 75235
Address all correspondence and requests for reprints to: Peter S. Johnston, Purdue Pharma, 100 Connecticut Avenue, Norwalk, Connecticut 06850.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
-glucosidase inhibitor miglitol
vs. the sulfonylurea glyburide in the treatment of
elderly patients with type 2 diabetes mellitus, inadequately controlled
by diet alone. This was a double-blind, randomized, placebo-controlled,
1-yr trial of miglitol 25 mg TID and 50 mg TID compared with placebo
and a titrated dose of glyburide in a parallel group comparison study
conducted in 30 outpatient sites across the United States. Four hundred
eleven (411) diet-treated patients age 60 yr or greater were randomized
to receive either placebo TID (n = 101), miglitol 25 mg TID
(n = 104), miglitol 50 mg TID (n = 102), or a once-daily dose
of glyburide titrated based on fasting plasma glucose (FPG) (n =
104), for a period of 56 weeks. Efficacy was assessed by glycated
hemoglobin (HbA1c), fasting and post-meal glucose, insulin, and lipid
levels, and by 24-h urinary excretion of glucose and albumin. Safety
and tolerability were assessed by tabulation of adverse events,
periodic laboratory determinations, and home blood glucose
monitoring. HbA1c treatment effects (placebo-subtracted change in HbA1c from baseline) at the 1-yr endpoint were -0.49%, -0.40%, and -0.92% in the miglitol 25 mg TID, miglitol 50 mg TID, and glyburide groups, respectively (P < 0.05 - 0.01 vs. placebo). Postprandial insulin levels were significantly greater than placebo and miglitol in the glyburide group (P < 0.01). Hypoglycemia, weight gain, and both routine and serious cardiovascular events were more frequent in the glyburide group (P < 0.05 - 0.01 vs. placebo or miglitol groups). Diarrhea (or soft stools) and flatulence were more common in both miglitol groups than in the other two groups in a dose-dependent manner, but resulted in relatively few study dropouts.
Treatment with miglitol offers the elderly type 2 diabetic patient
significant reductions in daylong glycemia as measured by
HbA1c. The greater HbA1c reductions seen with
once-a-day glyburide occurred at a cost of significant increases in
weight, insulin levels, and the incidences of clinical and subclinical
hypoglycemia, which did not occur in the miglitol groups.
-glucosidase inhibitors are a useful and relatively safe therapeutic
option in the elderly patient with type 2 diabetes.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Insulin injection is a technique that elderly patients may resist and may find difficult, and insulin administration has a well-known risk of hypoglycemia. Sulfonylurea (SFU) oral agents have a more insidious tendency to induce hypoglycemia because of their long duration of action, especially in elderly patients, and it may be that a dose of SFU sufficient to "normalize" fasting blood glucose exposes some elderly patients to an unacceptable risk of hypoglycemia at other times during the day (3, 4). In a study comparing glyburide and glipizide treatment of elderly type 2 patients with moderate hyperglycemia, a high incidence of asymptomatic, subclinical hypoglycemia (glucose levels 61–81 mg/dL) was detected by memory glucose meter monitoring (5). Symptomatic episodes tended to occur after unusual exertion or activity as opposed to overnight and showed no correlation with sulfonylurea dose. Biguanides, which probably cause less hypoglycemia as monotherapy than SFUs, have reduced clearance in the elderly, and may be risky in those with mild to moderate renal impairment, some of whom will have "normal" serum creatinine levels.
Hyperglycemia in the elderly may have a large postprandial component. Whereas average fasting glucose levels in the population as a whole increase by 1–2 mg/dL per decade with aging, the average 2-h postprandial level increases by 4–20 mg/dL over the same 10-yr period (2, 4, 6). While hepatic sensitivity to insulin (in terms of reducing endogenous glucose production) does not decrease with increasing age, peripheral tissue sensitivity to insulin [reflected in increased postprandial glucose concentrations and increased glycated hemoglobin (HbA1c) values] is reduced in the elderly (7, 8). Results of the Bedford Survey (9) revealed an increased incidence of coronary artery disease in borderline diabetic individuals and in those patients with the "hyperglycemia of aging," suggesting that decreased insulin sensitivity in the elderly is not a benign process.
-glucosidase inhibitors (GIs), which reduce or delay carbohydrate
digestion by competitive enzyme inhibition at the ciliated border of
the small intestine, may be an attractive therapeutic modality in
elderly type 2 diabetic patients. Because carbohydrate is digested and
absorbed more slowly, large postprandial increases in blood glucose and
serum insulin levels are prevented or reduced. Miglitol is an GI
recently approved in the United States and Europe for the treatment of
type 2 diabetes, in doses of 25, 50, and 100 mg TID. The 25 mg TID and
50 mg TID doses were chosen for this study based on the outcomes of US
Phase II NIDDM miglitol studies (10), which indicated similar efficacy
and a reduced incidence of gastrointestinal side effects at the 50 mg
dose compared with the 100 mg dose. It was thought that in the elderly
type 2 diabetic population miglitol 25 mg TID might generate
substantial efficacy with fewer adverse effects compared with higher
doses.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Patients with type 2 diabetes aged 60 yr old and above were recruited at 30 study centers across the U.S. At least 60% of the patients at each center were required to be age 65 or older. They were to have been treated with diet alone for their diabetes for at least 12 weeks before randomization at Visit 4. HbA1c levels were required to be from 6.5% to 10% inclusive (normal range: 4–6%), and fasting plasma glucose levels greater than 140 mg/dL at Visit 3, 2 weeks before randomization. Patients were required to be able to understand and comply with diet and glucose monitoring guidelines (see below) and to be free of serious illness that would prevent them from satisfactorily completing the 1-yr study.
Study design
A diagram of the study design is given in Fig. 1
. After an initial screening period,
patients entered a 6-week run-in phase, during which they received
placebo versions of both miglitol and glyburide in single-blind
fashion. They received instruction in blood glucose monitoring using
memory glucose meters (Glucometer M+®; Bayer Diagnostics,
Tarrytown, NY) and received instruction in an ADA-approved diet of at
least 50% carbohydrate, designed to promote a 1-pound-per-week weight
reduction for overweight patients. Eligibility for randomization was
determined by medical history and physical examination, and by chest
x-ray, electrocardiographic (ECG) and laboratory results early in
screening, as well as by laboratory findings (including
HbA1c) at Visit 3. Patients’ informed consent was obtained
at the first screening visit, in accordance with the guidelines of the
study centers’ ethical review boards, which had approved the study
protocol.
|
Patients randomized to miglitol 25 mg TID or 50 mg TID took their doses
with the first bite of each main meal and could not change their
miglitol dose for the duration of double-blind treatment. Patients
assigned to receive miglitol 50 mg TID were treated with miglitol 25 mg
TID for the first 2 weeks of double-blind treatment, as shown in Fig. 1
. The study concluded with the visit and
test meal at 56 weeks.
Measurements
The primary efficacy variable was change in HbA1c between randomization (baseline) and the 1-yr endpoint, defined as the last valid visit after Week 40. HbA1c levels were assayed periodically during the study to permit ongoing determinations of changes from baseline, and these intermediate HbA1c changes were secondary efficacy variables. Other secondary efficacy parameters included changes from baseline in plasma glucose and in serum insulin and triglyceride levels, both fasting and 60, 90, and 120 min after a standardized test meal (480 calories, 51% carbohydrate) given at baseline, after 6 months (24 weeks), and after 1 yr of treatment, or at premature termination. Fasting lipid levels (total, low density lipoprotein, and high density lipoprotein cholesterol) and changes in 24-h normalized urinary albumin and glucose excretion were also efficacy parameters. In addition, the incidence of treatment failures and treatment responders in each study group were examined. Treatment failure was defined either as initiation of chronic antihyperglycemic treatment other than that specified and permitted by the protocol, or as study termination resulting from unacceptable blood glucose control, as determined by the investigator. Treatment responders were patients who demonstrated a reduction in HbA1c from baseline to endpoint of at least 1.0%, or whose endpoint HbA1c was 7.0% or less, having fallen by at least 0.5% from baseline.
Central laboratories were used for the performance of efficacy and safety laboratory measurements during the study to minimize variability and to permit pooling of results across centers. HbA1c levels were assayed by the University of Missouri (Dr. David Goldstein), and all other blood and urine assays were performed by SmithKline Beecham Clinical Labs (Van Nuys, CA).
Safety and tolerability data primarily consisted of tabulations of adverse events (including intercurrent illnesses), changes in vital signs, and periodic laboratory determinations [complete blood counts, urinalyses, serum chemistries including liver function tests, serum levels of vitamins (such as folate, B12, 25-OH vitamin D, and B6), and iron and magnesium levels], as well as records of patients’ home glucose monitoring results, downloaded from patients’ memory meters into site computers at each visit.
Dietary compliance and changes in diet composition were assessed by 3-day diet diaries completed just before the baseline, 6-month, and 1-yr visits. Compliance with study medication (miglitol and glyburide) was evaluated by tablet counts.
Statistical analysis
All significance tests were two-tailed and were performed at an
-level of 0.05. The primary efficacy variable was the change from
baseline in HbA1c at the 1-yr endpoint.
Continuous efficacy variables were analyzed by analysis of variance
(ANOVA). The ANOVA model included effects for treatment and
investigator. The pairwise comparisons were based on the least-squares
means estimated by the model. The categorical efficacy variables were
analyzed using the Mantel-Haenzel test, stratifying by investigator.
The primary efficacy comparison was the test of difference between
miglitol 50 mg TID and placebo. All other pairwise comparisons were to
be performed secondarily. Incidence rates of adverse events and
abnormal laboratory values were analyzed by 
2 or Fisher
exact tests, depending on cell sizes. If the normal P value
for treatment differences was no more than 0.20, pairwise comparisons
were performed. Although there were many adverse events and labs to be
analyzed, no adjustment was to be made for the multiplicity of tests
because the P values were to be used as flags to highlight
differences of potential importance between treatment groups.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
presents demographic and
metabolic characteristics of the valid-for-efficacy patients in the
four treatment groups at baseline. As can be seen, there were no
differences between the groups at the onset of double-blind treatment.
Likewise, the baseline diets of the four groups were comparable in
terms of both percent composition and of overall caloric totals (based
on review of 3-day diet diaries) and remained so throughout the year of
study. The mean and median glyburide doses in the glyburide group at
the end of 1 yr of treatment were 8.9 mg/day and 3.75 mg/day,
respectively, for patients valid for efficacy. The median glyburide
dose in patients valid for safety (all randomized patients) was 5
mg/day.
|
Mean changes from baseline in HbA1c at the 1-yr
(primary) endpoint for each treatment group are presented in Table 2. Mean HbA1c levels remained
practically constant in the placebo group, but fell in each active
treatment group. The mean placebo-subtracted HbA1c
reduction from baseline (treatment effect) in the miglitol 25 mg group
was slightly greater than the corresponding value in the miglitol 50 mg
group, but the two treatment effects were not significantly different
from each other. The glyburide treatment group had a significantly
larger HbA1c decrease from baseline than either miglitol
group. As can be seen in Fig. 2, HbA1c
reductions were evident by 12 weeks of treatment and were maintained
over the year of treatment. Responder analysis, based on
HbA1c responses in the four treatment groups, demonstrated
a significantly greater percentage of patients in the miglitol 25 mg,
miglitol 50 mg, and glyburide groups who were valid for an
intent-to-treat analysis and who qualified as treatment responders
(33%, 30%, and 52%, respectively) than in the placebo group (18%)
(P < 0.05).
|
|
. All
three active treatments showed significant reductions in fasting plasma
glucose (FPG) and postprandial plasma glucose at each test meal
timepoint, and for the glucose AUC (area under the curve) measured over
the 2 h after the start of the meal. There were no significant
differences between the glucose treatment effects seen in the miglitol
50 mg group and in the glyburide group except at the 60-min
postprandial timepoint, at which time the reduction in the miglitol 50
mg group was significantly greater than in the glyburide group. Fig. 3A is a graph of the mean plasma glucose
changes from baseline at each test meal timepoint in each treatment
group, at the 1-yr endpoint. Though the reductions in the miglitol 50
mg group exceeded those in the miglitol 25 mg group at each timepoint,
these differences were not significant, nor were the glucose AUC
reductions in the three active treatment groups significantly
different, although the greatest AUC reduction occurred in the miglitol
50 mg group. Postprandial plasma glucose excursions above fasting were
reduced most in the miglitol groups, with very little change in the
glyburide group, over the year of treatment.
|
displays the mean change from baseline in serum insulin at the 1-yr
test meal timepoints for each treatment group. Fasting insulin levels
were essentially unchanged in either miglitol group, although
reductions from placebo were seen in the miglitol groups at the three
postprandial timepoints. These placebo-subtracted differences were
significant at 60 min for the miglitol 25 and 50 mg groups, and at 90
min for the 50 mg group. On the other hand, significant increases in
serum insulin were seen at the fasting and at all postprandial
timepoints in the glyburide group compared with each of the other
groups. Glyburide-associated postprandial insulin changes increased
progressively from fasting through 120 min after the meal. There were
no significant differences at any timepoint between the serum insulin
changes seen in the two miglitol groups.
Little change between treatment groups was evident in any fasting lipid
parameter, in postprandial triglyceride levels, or in the normalized
urinary excretion of glucose or albumin over the year of treatment
(Table 2). The incidence of treatment failure was significantly lower
in the glyburide group (2%) than in the other three groups (11%,
placebo and miglitol 25 mg, and 6%, miglitol 50 mg) (P
< 0.05).
Safety
All randomized patients were included in the safety
analysis. Table 3 lists those adverse events seen during
double-blind treatment that occurred significantly more often in one
treatment group than in another. These incidence rates reflect any
reporting of an event, no matter how mild or brief. Of these, diarrhea,
flatulence, nausea, and vomiting occurred more commonly among
miglitol-treated patients than in placebo- or glyburide-treated
individuals. Almost all of these events were mild to moderate in
severity, and they were often transient. Treatment-emergent episodes of
constipation occurred in 7% of placebo-and glyburide-treated patients,
but in only 2% and 5% of patients treated with miglitol 25 mg TID and
50 mg TID, respectively (not significant).
|
Results of home glucose monitoring performed before meals are shown in
Table 4. As
Table 4 shows, significant hypoglycemic readings
(i.e. less than 60 mg/dL) occurred as often in
the glyburide group (a total of 254 events) as in the other three
groups combined. These low readings were not attributable only to a
small subset of susceptible individuals, as meter readings less than 60
mg/dL occurred in 49% of glyburide-treated patients. This
preponderance of glyburide patients in each category of hypoglycemic
readings (<80 mg/dL and <60 mg/dL) occurred throughout the year
of treatment, though the greatest disproportion occurred during the
glyburide titration phase. Patients were almost always asymptomatic at
these times.
|
, cardiovascular adverse events as a whole were
significantly more common in the glyburide group than in the miglitol
50 mg group, both in terms of routine adverse events
(e.g. peripheral edema) and those of a more
serious nature, such as those which necessitated or prolonged
hospitalization, or caused death (e.g. stroke,
coronary artery disorder, which occurred in 3% of patients in the
glyburide group vs. 0% of miglitol 50 mg patients). The disproportionate occurrences between treatment groups of digestive surgery, hypoesthesia, and bronchitis are without obvious explanation.
Four study patients died during the year of double-blind treatment. Two glyburide patients died, one from a cerebrovascular accident and one from metastatic prostate cancer 43 days after the patient discontinued the study because of a cerebrovascular accident. One patient in the miglitol 25 mg group died from a stroke, and one patient randomized to the miglitol 50 mg group died because of metastatic lung cancer.
There were no significant differences between treatments in the rates of study discontinuation due to adverse events (6% placebo, 10% miglitol 25 mg, 12% miglitol 50 mg, and 6% glyburide). The most common reasons for study discontinuation were hyperglycemia, diarrhea, and flatulence in the miglitol groups, and cardiovascular adverse events in the glyburide group.
Mean body weight increased significantly and continuously in glyburide-treated patients over the course of the year of treatment. The mean increase from baseline to the last visit in the glyburide group was 2.3 kg (P = 0.0001 vs. placebo). Gradual decreases from baseline over the course of the study were seen in the placebo, miglitol 25 mg, and miglitol 50 mg groups (-1.1, -2.3, and -1.5 kg, respectively, at the last visit).
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
There was a large increase in the number of reported hypoglycemic episodes in the glyburide group, compared with any of the other treatment groups. These were clinical episodes judged by the investigator to likely be caused by low blood glucose, based on reporting of patients’ symptoms, recent blood glucose or HbA1c values, relation of symptoms to food intake and activity level, and any glucose monitoring results collected at or close to the time of symptoms. Glyburide dose titration was performed in a more cautious manner than generally occurs in clinical practice to minimize the risk of severe hypoglycemia. Doses were increased through as many as 7 levels over 14 weeks, allowing 2 weeks at each dose stage, with frequent home monitoring of blood glucose and corroboration by laboratory FPG measurements at each dose level. Despite these precautions, there was still a large excess of symptomatic hypoglycemia in patients in the glyburide group (88% of all episodes) throughout the year of study. These primarily occurred after breakfast and lunch rather than overnight.
The relatively low median glyburide dose in the valid-for-safety glyburide treatment group after the titration period (5 mg/day) is consistent with other studies of sulfonylurea-induced hypoglycemia in elderly type 2 diabetic patients (11), in which serious morbidity related to hypoglycemia is as equally likely to occur at low doses of medication as at high doses, even of second-generation agents. The diminished perception of hypoglycemia by elderly type 2 diabetic patients and a supposedly increased risk of stroke or myocardial infarction accompanying hypoglycemic episodes in the elderly (12) are further reasons for physicians to exercise caution in prescribing hypoglycemic drugs in this population. Actual hypoglycemia may have been more common than the data indicate, because of the relative insensitivity of older type 2 patients to the symptoms of hypoglycemia (13).
The similarity between the placebo group and the two miglitol groups in
the incidence of hypoglycemia is consistent with other miglitol trials,
in particular with a 1-yr trial of miglitol 50 mg TID alone, compared
with a fixed dose of glyburide (2.5 mg BID), compared with the
combination of the two regimens, in diet-treated type 2 patients. In
that trial, rates of hypoglycemia in the placebo (9%), miglitol 50 mg
TID (5%), and glyburide-alone (40%) groups were similar to rates in
this study. The combination of miglitol and glyburide in that study was
associated with a lower incidence of hypoglycemia (36%) than in the
glyburide-alone arm (not significant). The present study’s results are
more impressive given that investigators could reduce glyburide (but
not miglitol) doses when hypoglycemia occurred or threatened, unlike
the earlier study. There may be a tendency for GIs to protect
against the post-meal hypoglycemic effects of certain pharmacologic
agents (such as SFU), as the absorption of meal-derived nutrients is
prolonged with GI treatment.
Results of home blood glucose monitoring followed a similar pattern. The percents of patients with glucose readings either less than 80 mg/dL or less than 60 mg/dL were prominently greater in the glyburide group than in the other treatment groups. Although one might have expected a greater number of readings to have been performed by patients in the glyburide group because of their lower mean ambient blood glucose values and their greater number of hypoglycemic events, there was not a notable difference between the total number of readings done by patients in the glyburide group and in those in the other treatment groups, including those patients with a greater proportion of readings less than 100 mg/dL.
A review of the incidence of adverse events (AEs) by body system reveals a significantly greater incidence of cardiovascular AEs in the glyburide group than in the miglitol 50 mg treatment group. Although there was no individual cardiovascular AE that occurred significantly more commonly in the glyburide group (other than peripheral edema), the overall incidence of cardiovascular events and of serious cardiovascular AEs was significantly greater in the glyburide group compared with the miglitol 50 mg TID group, and the increased incidence of peripheral edema in the glyburide group may be reflective of this. Although the possible association between sulfonylurea treatment of type 2 diabetes and increased cardiovascular morbidity that was observed in the UGDP study of the 1960’s (14) has generally not been confirmed in subsequent studies, elderly patients, already at increased risk for cardiovascular disease because of their age, may be particularly susceptible to any potential adverse cardiovascular effects of SFU agents.
The incidences of the gastrointestinal adverse events in the miglitol
groups were commensurate with what has been seen in other long-term
trials of miglitol at the doses used. For flatulence there was less of
a difference between the incidences in the placebo group and the two
miglitol groups than was seen in two other long-term U.S. fixed-dose
studies of miglitol (10, 15). These symptoms are the result of
intestinal -glucosidase inhibition, which results in the appearance
of unabsorbed carbohydrate in the distal small bowel and colon, where
its metabolism by microflora produces the gas that results in
flatulence, and where its osmotic effect gives rise to softer and more
frequent stools (more commonly than actual watery diarrhea). As in
other studies of the drug, no serious morbidity resulted from these
adverse events, and they resolved promptly in the (relatively) rare
instances where drug discontinuation because of them was required
(1–3% of patients in the miglitol groups). Nausea and vomiting have
not been seen more commonly in miglitol-treated patients in other
long-term trials of the drug (10, 15, 16, 17).
The percentages of patients reporting diarrhea and flatulence in the
miglitol groups tended to fall over the year of the study, though the
percentages of patients reporting diarrhea and flatulence in the
miglitol 50 mg group remained greater than in the placebo group
throughout the year. Although the incidence of treatment-emergent
episodes of constipation were low in this study, they tended to be
lower in the miglitol groups. The tendency of GIs to induce soft or
loose stools, and consequently to result in relief from constipation,
may benefit many elderly patients.
Body weight increased from baseline to the last visit by 2.3 kg in the
glyburide group and fell by 1.1 to 2.3 kg in the other groups. These
weight changes began to be evident early in the study and continued
throughout the course of treatment. Weight gain, as noted above, is
common with SFU treatment, however, GIs have had no significant
effect on body weight compared with placebo in any other long-term
trials (10, 15, 16, 17). The increased blood insulin levels seen with SFU
treatment may have contributed to the associated weight gain, and
hyperinsulinemia alone or in combination with weight gain may
contribute to any excess cardiovascular morbidity or mortality
associated with sulfonylurea treatment (18).
HbA1c treatment effects in the glyburide group were significantly superior to placebo and to both miglitol treatment groups, although titration of glyburide was permitted to optimize glycemic control in this arm and was not permitted for miglitol at either dose. Examination of the changes from baseline in HbA1c by cofactors such as age, gender, duration of diabetes, body mass index, baseline HbA1c, baseline FPG, baseline 1-h test meal glucose and insulin levels, or percent dietary carbohydrate at baseline, revealed little relationship between HbA1c treatment effects and these variables.
Whereas HbA1c treatment effects in the miglitol 25 mg group were greater in this study than in other placebo-controlled trials of miglitol (15), the HbA1c treatment effects in the miglitol 50 mg group seem incommensurate with the corresponding placebo-subtracted changes in test meal blood glucose levels in that group, based on relationships between test meal effects and HbA1c effects of this dose in other long-term studies (10, 15). Compliance issues may have played a role in the HbA1c treatment responses in the miglitol 50 mg group, though mean compliance with medication regimens based on tablet counts was 98–100% at all study visits for all groups. In any event, the HbA1c results in this study in the 50 mg TID group stand in marked contrast to those of a placebo-controlled 16-week miglitol study performed in Finland in elderly (ages 65–80 yr) vs. younger (ages 30–60 yr) type 2 diabetic patients, in which an HbA1 treatment effect of 1.44% was seen in the older patients, vs. 0.41% in younger patients (data on file, Bayer Corporation). That study used the 100 mg TID dose of miglitol, but in the U.S. fixed-dose miglitol studies there has not been a significant difference between glycated hemoglobin treatment effects at the 50 mg TID and 100 mg TID doses of miglitol (10, 15).
As expected, mean FPG levels were reduced most in the glyburide group,
but not statistically significantly more than in the miglitol 50 mg
group. Glyburide’s glucose-lowering effect was confined to FPG
reduction, with essentially no impact on postprandial glucose
excursions, at which time effects of miglitol were prominent (Fig. 3A).
Fasting glucose has a quantitatively greater effect on the mean daylong
blood glucose level than have the postprandial periods; consequently,
the overall impact of glyburide on HbA1c was greater than
that of miglitol.
Changes in serum insulin levels in the active treatment arms were
consistent with both drugs’ known mechanisms of action. Serum insulin
increases above baseline in the glyburide group during the test meal at
study endpoint were greater at later test meal timepoints (Fig. 3B),
but resulted in no additional postprandial glucose lowering in that
group (Fig. 3A). Increases in daylong insulinemia may have an adverse
bearing on cardiovascular risk profiles of type 2 patients, as other
work has suggested (9, 18), and insulin secretagogues may accelerate
ß-cell exhaustion in type 2 diabetes (19).
Despite the changes in glucose and insulin, lipid parameters showed little change from baseline in any treatment group. Urinary glucose and albumin excretion values at the 1-yr endpoint were comparable to baseline levels in all treatment groups as well. Age-related decrements in renal function in the elderly, and a consequently higher threshold for renal glucose excretion than in younger patients, may be partly responsible for the urine glucose results in this study. The similarity between miglitol’s and glyburide’s effects on urinary glucose excretion despite the disparity in HbA1c effects of the two treatments is testimony to the relatively large contribution of postprandial glucose excursions to urinary glucose losses in diabetes. It also offers evidence that the weight increases seen in glyburide-treated patients were not simply caused by reduced urinary glucose losses, as weight gain did not occur in the miglitol groups despite similar effects on 24-h urine glucose. Some of the weight gain in the glyburide group may be attributable to the relative hyperinsulinemia experienced by these patients and/or to a pressure for increased caloric intake driven by (relative or absolute) hypoglycemia.
Miglitol treatment offers the elderly patient with mild type 2 diabetes
significant improvement in glycemic control, especially postprandially,
at a cost of mild to moderate gastrointestinal side effects that are
relatively harmless and tend to abate with continued use of the drug.
GIs may be superior in this regard to sulfonylureas such as
glyburide, which when titrated to achieve desirable control of fasting
plasma glucose levels, exposes patients to an enhanced risk of
hypoglycemia, hyperinsulinemia, weight gain, and possibly to adverse
cardiovascular events.
|
Acknowledgments |
|---|
Paul Casner, MD, El Paso TX; Ken Cathcart, DO, Spokane, WA; Steven Dorfman, MD, Dallas TX; Steven Edelman, MD, San Diego CA; Daniel Einhorn, MD, San Diego CA; Gary Enzman, MD, Seattle, WA; Alan Garber, MD, Houston, TX; Ronald Goldberg, MD, Miami FL; George Grunberger, MD, Detroit, MI; Irl Hirsch, MD, Seattle WA; Priscilla Hollander, MD, St. Louis Park MN; Sherman Holvey, MD, Los Angeles, CA; Roy Kaplan, MD, Concord CA; Arthur Krosnick, MD, Princeton NJ; Edward LaCava, MD, Kirkland WA; Peter Lodewick, MD, Birmingham AL; Charles Lucas, MD, Birmingham MI; Alan Marcus, MD, Laguna Hills CA; Sergio Mather, MD, Fort Myers FL; Sam Miller, MD, San Antonio TX; David Munoz, MD, Tacoma WA; Catherine Niewoehner, MD, Minneapolis MN; Charles Peterson, MD, Santa Barbara CA; Sanford Plevin, MD, Palm Harbor FL; Philip Raskin, MD, Dallas TX; Sidney Rosenblatt, MD, Irvine CA; David Schimel, MD, Lake Bluff IL; Louis Shane, MD, White Plains NY; Donald Simonson, MD, Boston MA; Stuart Weiss, MD, San Diego CA.
Received July 17, 1997.
Accepted February 17, 1997.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
-glucosidase inhibitor miglitol (BAY m
1099) in NIDDM patients treated with maximum-dose sulfonylurea.
Diabetes 45(Suppl 2):221A.
-glucosidase inhibition. Diabetes
46(Suppl 1):156A.
-glucosidase
inhibition. Diabetes 46(Suppl 1):156A.
This article has been cited by other articles:
 |
|
 |
|
  W.S. Leslie, C.R. Hankey, and M.E.J. Lean Weight gain as an adverse effect of some commonly prescribed drugs: a systematic review QJM, July 1, 2007; 100(7): 395 - 404. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Lada and U. Idrees Toxicity of Oral Agents Used to Treat Diabetes Journal of Pharmacy Practice, June 1, 2005; 18(3): 145 - 156. [Abstract] [PDF]
|
|
|
|
 |
|
 F. A. van de Laar, P. L. Lucassen, R. P. Akkermans, E. H. van de Lisdonk, G. E. Rutten, and C. van Weel {alpha}-Glucosidase Inhibitors for Patients With Type 2 Diabetes: Results from a Cochrane systematic review and meta-analysis Diabetes Care, January 1, 2005; 28(1): 154 - 163. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. S. Meneilly and D. Tessier Diabetes in Elderly Adults J. Gerontol. A Biol. Sci. Med. Sci., January 1, 2001; 56(1): 5M - 13. [Abstract] [Full Text]
|
|
|
|
 |
|
 S. Matthaei, M. Stumvoll, M. Kellerer, and H.-U. Häring Pathophysiology and Pharmacological Treatment of Insulin Resistance Endocr. Rev., December 1, 2000; 21(6): 585 - 618. [Abstract] [Full Text]
|
|
|
|
|
|
J. E. Morley Editorial: Diabetes Mellitus: A Major Disease Of Older Persons J. Gerontol. A Biol. Sci. Med. Sci., May 1, 2000; 55(5): 255M - 256. [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
