OBJECTIVES: To review the clinical pharmacology of acarbose, an alpha-
glucosidase inhibitor, and to summarize its role in the pharmacotherap
y of diabetes mellitus. DATA SOURCES: A MEDLINE search identified all
relevant articles, including reviews; Bayer Pharmaceuticals. STUDY SEL
ECTION: Due to the large number of clinical trials available, specific
criteria were used to narrow the focus of this review: (1) randomized
, double-blind, placebo-controlled, parallel-group study design; (2) a
minimum of 25 patients enrolled per treatment arm; (3) a treatment du
ration of 90 days or more; and (4) adherence to Food and Drug Administ
ration Good Clinical Practice guidelines. DATA EXTRACTION: All clinica
l trials that were available up to December 1995 were reviewed. Prelim
inary trials and unpublished reports were not reviewed. DATA SYNTHESIS
: Acarbose is effective in reducing postprandial hyperglycemia. It doe
s not stimulate endogenous insulin secretion and, therefore, will not
cause hypoglycemia when used as monotherapy. The enhanced glycemic con
trol achieved with acarbose is additive to that of sulfonylureas. It l
owers postprandial serum glucose and insulin concentrations and does n
ot promote weight gain. Acarbose can be used as first-line therapy wit
h diet and exercise, or it can be used in combination with sulfonylure
as to lower hemoglobin A(1c) concentrations an additional 0.5-0.9%. Ac
arbose is not a cure for diabetes, nor is it a substitute for diet, ex
ercise, oral hypoglycemic agents, or insulin. Adverse effects are gast
rointestinal and can be diminished by starting with an initial dosage
of 25 mg tid. Depending on patient response, the dosage can be increas
ed up to a maximum of 100 mg tid over time. CONCLUSIONS: Acarbose, thr
ough its unique mechanism of action, appears to be a safe and effectiv
e adjunctive agent to diet/exercise therapy or sulfonylurea therapy fo
r treatment of non-insulin-dependent diabetes mellitus.