STOP-NIDDM

Published

Chiasson JL, et al. "Acarbose for prevention of type 2 diabetes mellitus". The Lancet. 2005. 359(9323):2072-2077.
PubMed • Full text • PDF

Clinical Question

In patients with impaired glucose tolerance, does acarbose reduce the risk of progressing to type 2 diabetes mellitus (T2DM), as compared to placebo?

Bottom Line

In patients with impaired glucose tolerance, acarbose reduces the risk of progressing to T2DM, as compared to placebo. The most common adverse effects were flatulence and diarrhea.

Major Points

Impaired glucose tolerance is associated with a significant risk of progressing to type 2 diabetes mellitus (T2DM). Alpha-glucosidase inhibitors (eg, acarbose) are reversible inhibitors of alpha-glucosidase, an enzyme located on the brush border of the small intestine. The net effect is a delay in absorption of carbohydrates thus lowering postprandial glucose levels.

The Study to Prevent NIDDM (STOP-NIDDM) randomized 1,429 patients with impaired glucose tolerance to receive acarbose or placebo. The primary endpoint was the incidence of diabetes. Acarbose significantly reduced the risk of the primary endpoint versus placebo (32% vs. 42%; relative risk 0.75, 95% CI 0.63-0.90, P=0.0015). However, up to a quarter of patients discontinued treatment early. In the acarbose group, 13% of treatment discontinuation occurred due to gastrointestinal effects, mainly flatulence and diarrhea.

In a separate publication, the authors reported the beneficial effect of acarbose on cardiovascular (CV) risk. Acarbose significantly reduced the relative risk of CV events (coronary heart disease, CV mortality, heart failure, cerebrovascular event, and peripheral vascular disease) by 49% (hazard ratio [HR] 0.51, 95% CI 0.28-0.95, P=0.03). The risk of hypertension was also reduced (HR 0.66, 95% CI 0.49-0.89, P=0.006).

T2DM confers significant risk for multiple chronic disease, including CVD. A 2018 Cochrane review assessed the efficacy of alpha-glucosidase inhibitors for prevention of T2DM and T2DM complications. There was ARR of ~4% of incident T2DM among 3 trials (16.7% vs. 20.3%; RR 0.82; P<0.001). However, there was no reduction in events from T2DM, including CVD, stroke, HF, and death with this intervention from relatively low-quality evidence. Given a lack of reduction in these T2DM-related conditions, the clinical utility of alpha-glucosidase inhibitors among patients at high risk of T2DM development remains unresolved.

Guidelines

ADA Standards of Medical Care in Diabetes (2023, adapted)

• Pharmacotherapy (eg, for weight management, minimizing the progression of hyperglycemia, cardiovascular risk reduction) may be considered to support person-centered care goals (grade B).

Design

• Double-blind, randomized, placebo-controlled trial
• N=1,429 patients with impaired glucose tolerance
  • Acarbose (n=714)
  • Placebo (n=715)
• Setting: 9 countries
• Enrollment: 1995-1998
• Mean follow-up: 3.3 years
• Analysis: intention-to-treat
• Primary outcome: T2DM

Population

Inclusion Criteria

Published elsewhere

• Impaired glucose tolerance, defined as
  • 2-h plasma glucose concentration 7·8-11 mmol/L after a 75 g glucose load
  • fasting plasma glucose concentration of 5·6-7·7 mmol/L
• BMI 25-40 kg/m²

Exclusion Criteria

• DM
• serum creatinine ≥130 µmol/L (1.5 mg/Ll)
• fasting serum triglyceride ≥10 mmol/L >886 mg/dL)
• liver enzymes >1.8 times the upper limit of normal (ULN)
• TSH >1.5 times above the ULN or below the lower limit of normal (<0.3 mU/L)
• systemic glucocorticoids, ß-blockers, thiazide diuretics, and nicotinic acid treatment within last 3 months
• use of drugs likely to impair intestinal motility or absorption
• recent cardiovascular event

Baseline Characteristics

From the acarbose group

• Demographics: age 54.3 years, 48% male, 97% Caucasian
• BMI: 31±4.3 kg/m² ; waist circumference: 102.1±11.7 cm
• Fasting plasma glucose: 6.23±0.5 mmol/L; 2-hour OGTT 9.26±1.06 mmol/L
• BP: systolic 131.4±16.3 mm Hg, diastolic 82.8±9.4 mm Hg
• LDL-C :3.66 mmol/L; HDL-C: 1·19 mmol/L; triglyceride: 2.07 mmol/L; total cholesterol 5.76 mmol/L
• Other medical history: active smoker 12%

Interventions

• Participants were randomized to acarbose or placebo in addition to their existing medications
• Acarbose was initiated at 50mg mane and increased as tolerated to the maximum of 100 mg three times a day
• Participants were reviewed every 6 months by the trial investigators
• At the end of treatment, outcome measures were repeated after following all patients who had not developed diabetes on placebo in a single-blind fashion for 3 months

Outcomes

Comparisons are acarbose vs placebo

Primary outcomes

Diabetes mellitus

32.4% vs. 41.5% (RR 0·75, 95% CI 0·63-0.9, P=0·0015)

Other outcomes

Reversion to normal glucose tolerance

35% vs. 31% (P<0.0001)

Change in weight

-0.5 kg vs. +0.3 kg (difference 0·77 kg, 95% CI 0·01-1·54, P=0·0184)

Subgroup Analysis

• There were no significant interactions in pre-specified subgroup analyses (age, sex, BMI).

Adverse events

Gastrointestinal events: 83% vs 60% (P<0.0001)

• leading to discontinuation: 13% vs 3%
• flatulence: 68% vs 27%
• diarrhea: 32% vs 17%
• abdominal pain: 17% vs 12%
• dyspepsia: 7% vs 9%
• constipation: 4% vs 5%

Cardiovascular events: 31% vs. 40% (P-value not given)

Respiratory events: 32% vs. 39% (P-value not given)

Musculoskeletal events: 34% vs. 39% (P-value not given)

Metabolic and nutritional events: 31% vs. 32% (P-value not given)

Nervous system events: 27% vs. 31% (P-value not given)

Urogenital events: 25% vs. 28% (P-value not given)

Skin events: 21% vs. 24% (P-value not given)

Haematological and lymphatic events: 4% vs. 6% (P-value not given)

Endocrine events: 4% vs. 4% (P-value not given)

Criticisms

• It was unclear if the beneficial effects of acarbose seen in the trial were due to prevention, masking effect or delaying of the onset of DM.
• The changes in blood glucose level and HbA_1c were not reported in the study.

Funding

• Bayer AG