The aim of this study was to determine if differing periods of prior hyperinsulinemic nonhypoglycemia can modify the subsequent counterregulatory response to hypoglycemia. Experiments were carried out on 19 normal 18-h fasted conscious dogs. Insulin was infused intraportally at 8 mU.kg-1.min-1 for 3 h on two occasions and 3.5 h on a third separate occasion. This resulted in similar steady-state arterial insulin levels during each protocol (4,370 +/- 433 pmol/1). Each animal was maintained at a similar plasma glucose nadir (2.8 +/- 0.6 mmol/1) for 2 or 2.5 h, depending on the protocol. In protocol I (n = 7) plasma glucose was allowed to fall to the desired hypoglycemic plateau by 30 min. In a second group of dogs (protocol II, n = 5) there was a 30-min period of euglycemic hyperinsulinemia followed by a 30-min fall (similar to protocol I) in plasma glucose. In a third group of dogs (protocol III, n = 7), there was an initial 15-min period of euglycemic hyperinsulinemia followed by a 45-min fall in plasma glucose. Differing periods of euglycemic hyperinsulinemia had distinct effects on subsequent counterregulation. During the final 2 h of hypoglycemia the incremental area under the curve (AUC) for glucagon was significantly greater in protocol I vs. II (3.0 +/- 1.0, -0.5 +/- 0.2 mug.l-1.min-1, P < 0.02, respectively). Conversely, catecholamine levels were increased in protocol II (30 min prior hyperinsulinemic euglycemia) compared with protocol I (epinephrine 1,448 +/- 268, 855 +/- 119 nmol.l-1.min-1; norepinephrine 244 +/- 30, 166 +/- 23 nmol.l-1.min-1, respectively, P < 0.05). During protocol III, glucagon and catecholamine levels were intermediate between protocols I (no euglycemic hyperinsulinemia) and II (30 min euglycemic hyperinsulinemia). AUC for hepatic glucose production (HGP), measured using [3-H-3]glucose or net hepatic glucose balance, was similar in each protocol. We conclude that 1) the duration of prior nonhypoglycemic hyperinsulinemia appears to play an important role in the counterregulatory response by subsequently suppressing glucagon release and 2) increased catecholamines can compensate for the blunted glucagon response and maintain HGP.