RT Journal A1 Brancati FL, Queale WS T1 A novel insulin sliding scale-reply JF Archives of Internal Medicine JO Archives of Internal Medicine YR 1997 FD November 24 VO 157 IS 21 SP 2524 OP 2524 DO 10.1001/archinte.1997.00440420162019 UL http://dx.doi.org/10.1001/archinte.1997.00440420162019 AB Trochtenberg suggests that sliding scale insulin regimens that reach a maximum of 20 U of insulin may perform better than those that stop at 12 U. Because the physicians in our study' generally did not prescribe such aggressive sliding scale insulin, we are unable to test this hypothesis directly. Nonetheless, we question this hypothesis for several reasons. First, there was no apparent association in our study between the type of sliding scale and the performance: aggressive scales performed as poorly as conservative scales. Second, the use of very aggressive sliding scale insulin should only increase the risk for hypoglycemia. Third, the beneficial effects of very aggressive sliding scale insulin on hyperglycemia are likely to be short-lived unless the underlying long-acting hypoglycemic regimen is modified accordingly. In light of Trochtenberg's comment that he "rarely treat[s] patients with diabetes mellitus using the sliding scale regimen alone," we suspect that the apparent success