Over 50% of patients with the polycystic ovary syndrome (PCOS) demonstrate excess levels of adrenal androgens (AAs), particularly dehydroepiandrosterone sulfate (DHS). Nonetheless, the mechanism for the AA excess remains unclear. It has been noted that in PCOS the pituitary and ovarian responses to their respective trophic factors (i.e. GnRH and LH, respectively) are exaggerated. Similarly, we have postulated that excess AAs in PCOS arises from dysfunction of the hypothalamic-pituitary-adrenal axis, due to 1) exaggerated pituitary secretion of ACTH in response to hypothalamic CRH, 2) excess sensitivity/responsivity of AAs to ACTH stimulation, or 3) both. To test this hypothesis we studied 12 PCOS patients with AA excess (HIDHS; DHS, >8.1 μmol/L or 3000 ng/mL), 12 PCOS patients without AA excess (LO-DHS; DHS, <7.5 μmol/L or 2750 ng/mL), and 11 controls (normal subjects). Each subject underwent an acute 90-min ovine CRH stimulation test (1 μg/kg) and an 8-h incremental iv stimulation with ACTH-(1-24) at doses ranging from 20-2880 ng/1.5 m2 · h) with a final bolus of 0.25 mg. All patient groups had similar mean body mass indexes and ages, and both tests were performed in the morning during the follicular phase (days 3-10) of the same menstrual cycle, separated by 48-96 h. During the acute ovine CRH stimulation test, no significant differences in the net maximal response (i.e. change from baseline to peak level) for ACTH, dehydroepiandrosterone (DHA), androstenedione (A4), or cortisol (F) or for the DHA/ACTH, A4/ACTH, or F/ACTH ratios was observed. Nonetheless, the net response of DHA/F and the areas under the curve (AUCs) for DHA and DHA/F indicated a greater response for HI-DHS vs. LODHS or normal subjects. The AUC for A4 and A4/F and the ΔA4/ΔF ratio (Δ = net maximum change) indicated that HI-DHS and LO-DHS had similar responses, which were greater than that of the normal subjects, although the difference between LO-DHS patients and normal subjects reached significance only for the AUC of the A4 response. No difference in the sensitivity (i.e. threshold or minimal stimulatory dose) to ACTH was noted between the groups for any of the steroids measured. Nonetheless, the average dose of ACTH-(1-24) required for a threshold response was higher for DHA than for F and A4 in all groups. No difference in mean responsivity (slope of response to incremental ACTH stimulation) was observed for DHA and F between study groups, whereas the responsivity of A4 was higher in HI-DHS patients than in normal or LO-DHS women. The net maximal and the overall (i.e. AUC) responses of DHA were greater for HI-DHS than for normal or LO-DHS women. The response of A4 and the ΔA4/ΔF ratio were greater for HI-DHS patients than for LO-DHS patients or normal subjects. Alternatively, HI-DHS and LO-DHS patients had similar overall responses (i.e. AUC) for A4 or A4/F, although both were greater than those of normal subjects. The relative differences in response to incremental ACTH stimulation between steroids was consistent for all subject groups studied, i.e. A4 > F or DHA. In conclusion, our data suggest that AA excess in PCOS patients is related to an exaggerated secretory response of the adrenal cortex for DHA and A4, but not to an altered pituitary responsivity to CRH or to increased sensitivity of these AAs to ACTH stimulation. Whether the increased responsivity to ACTH for these steroids is secondary to increased zonae reticularis mass or to differences in P450c17α activity, particularly of the Δ4 pathway, remains to be determined.