I am not convinced by Stenman and colleagues [1] that there is no benefit in increasing the dose of glipizide. Although blood glucose concentrations for the three treatment groups are not statistically different, the average daily blood glucose level is inversely proportional to the glipizide dose: 171 mg/dL for the group receiving 10 mg/d, 166 mg/dL for the group receiving 20 mg/d, and 160 mg/dL for the group receiving 40 mg/d. The mean hemoglobin A_1c (HbA_1c) values for the treatment groups are also not statistically different, but they may lack the sensitivity to detect small but statistically significant changes in blood glucose.

The authors do not specify on which days the twice-weekly home glucose measurements were obtained. If the measurements were discretionary, this may be a confounding factor in that patients may have recorded their sugars when most compliant. Moreover, because blood sugars were self-monitored, patients were aware of any difference in their glycemic control and may have altered their diet accordingly. Interestingly, some patients had normal baseline HbA_1c values yet experienced no hypoglycemia while receiving maximal doses of glipizide for 3 months.

The authors conclude that ß-cell function decreases in patients receiving higher doses of glipizide by measuring serum insulin levels after a test meal. However, enhancement of insulin secretion by sulfonylureas varies inversely with the underlying glucose concentration. Thus, if the baseline glucose level is lower, less insulin secretion by the ß-cells would be expected for a given glucose challenge.

Finally, the study group was small and heterogeneous and, according to the authors, individual responses to glipizide differed markedly. Given these issues, it might have been more useful to present the outcomes for all patients individually. Because results were presented as a group average, significant responses in some patients may have been diminished when combined with those patients who did not benefit from an increased glipizide dose.