Current issues of ACP Journal Club are published in Annals of Internal Medicine


Bedtime insulin plus metformin was effective and did not cause weight gain in patients with poorly controlled type 2 diabetes mellitus

ACP J Club. 1999 July-Aug;131:2. doi:10.7326/ACPJC-1999-131-1-002

Source Citation

Yki-Järvinen H, Ryysy L, Nikkilä K, et al. Comparison of bedtime insulin regimens in patients with type 2 diabetes mellitus. A randomized, controlled trial. Ann Intern Med. 1999 Mar 2;130:389-96.



Are there differences between 4 bedtime insulin plus oral hypoglycemic drug regimens in patients with poorly controlled type 2 diabetes mellitus receiving sulfonylurea monotherapy?


Randomized controlled trial with 1-year follow-up.


4 outpatient clinics at tertiary care hospitals in Finland.


96 patients with type 2 diabetes (mean age 58 y, 61% men, mean body mass index [BMI] 29 kg/m2) that was poorly controlled by sulfonylurea therapy alone. Inclusion criteria were age between 40 and 70 years, BMI < 35 kg/m2, fasting plasma glucose [FPG] level > 8 mmol/L (> 144 mg/dL), diabetes for > 3 years, previous glipizide or glyburide therapy, and fasting serum C-peptide level > 0.33 mmol/L (> 0.99 ng/mL). Exclusion criteria were heart failure, myocardial infarction, or stroke in the past 6 months; epilepsy or other severe disease; liver disease, elevated serum creatinine level, or macroalbuminuria; proliferative retinopathy or maculopathy; previous insulin therapy; excessive alcohol use; or night work. Follow-up was 92%.


Patients were allocated to bedtime intermediate-acting insulin, 100 IU/mL, plus glyburide, 10.5 mg, and metformin placebo; metformin, 2 g, and glyburide placebo; glyburide, 10.5 mg, and metformin, 2 g; or a second injection of insulin in the morning (24 patients/group). Patients self-adjusted their insulin to reach an FPG level < 6 mmol/L (< 108 mg/dL).

Main outcome measures

Body weight, symptomatic and biochemical (FPG < 3.5 mmol/L) hypoglycemic episodes, and measures of glycemic control.

Main results

At 1 year, the mean increase in body weight from baseline was not significant for patients treated with bedtime insulin plus metformin (0.9 kg) and was smaller than the mean increases for those treated with bedtime insulin plus glyburide (3.9 kg), bedtime insulin plus metformin and glyburide (3.6 kg), or bedtime and morning insulin (4.6 kg, P < 0.05). Patients treated with bedtime insulin plus metformin had fewer symptomatic hypoglycemic episodes (mean 1.8/patient) than did those treated with bedtime and morning insulin (mean 3.9/patient, P < 0.05). Mean glycosylated hemoglobin levels decreased from baseline by 2.5% (absolute change) with bedtime insulin plus metformin (P < 0.001); this decrease was greater than the decrease seen with the other regimens (P < 0.05).


After 1 year, patients with type 2 diabetes mellitus who were treated with bedtime insulin plus metformin did not gain weight, had fewer hypoglycemic episodes, and had better glycemic control than those treated with bedtime insulin plus glyburide, metformin and glyburide, or morning insulin.

Sources of funding: Academy of Finland; Orion, Espoo, Finland.

For correspondence: Dr. H. Yki-Järvinen, Department of Medicine, University of Helsinki, Haartmaninkatu 4, Box 340, FIN-00029 HUCH, Helsinki, Finland. FAX 358-9-471-2250.


Yki-Järvinen and colleagues show the benefits of combination therapy in patients with type 2 diabetes that was poorly controlled with sulfonylureas. The study raises an interesting question: Should insulin be used as a second step in these patients? Other studies have found that adding metformin to sulfonylureas improves diabetes control (1); however, the duration of such an improvement seems limited (2). In the United Kingdom Prospective Diabetes Study 33 (UKPDS 33), a gradual and progressive worsening of diabetes control was seen (2).

In this 1-year study by Yki-Järvinen and colleagues, blood glucose control was excellent in the group treated with metformin and bedtime insulin, but it is possible that patients will not maintain the same degree of glycemic control beyond 1 year and might require the addition of other oral agents or the use of multiple daily insulin injections. Nevertheless, this study provides insight into a practical, safe, and effective method for adjusting insulin by self-monitoring blood glucose levels.

To achieve continuing diabetes control in type 2 diabetes, combination therapy using oral agents or oral agents with insulin is often needed. Many patients with type 2 diabetes will eventually need to have insulin added to their treatment regimens because of β-cell failure. Insulin treatment is safe (associated with few hypoglycemic episodes) (2), effective, and not limited by a maximum dosage. Moreover, the evidence of the UKPDS does not support the notion that insulin treatment enhances arteriosclerosis (2).

Amir K. Hanna, MB
University of TorontoToronto, Ontario, Canada


1. DeFronzo RA, Goodman AM. Efficacy of metformin in patients with non-insulin-dependent diabetes mellitus. The Multicenter Metformin Study Group. N Engl J Med. 1995;333:541-9.

2. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet. 1998;352:837-53.