Cushing's disease (CD) is associated with increased morbidity and mortality. Until now, no medical treatment has been shown to be totally satisfactory when administrated alone. This study aimed to assess the effectiveness of cabergoline with added ketoconazole and of the same combination in reverse, using urinary free cortisol (UFC) and late night salivary cortisol (LNSC) levels as biochemical markers of the treatments' efficacy in CD patients. A prospective analysis conducted on 14 patients (f/m = 12/2; median age 52, range 33-70 years) divided into two groups: 6 patients initially treated with cabergoline for 4-6 months (rising from 0.5-1 mg/week up to 3.0 mg/week), after which ketoconazole was added (group A); and 8 patients first took ketoconazole alone for 4-6 months (rising from 200 mg/day to 600 mg/day), then cabergoline was added (group B). Patients were compared with 14 age-matched patients in prolonged remission after effective neurosurgery for CD. The combination therapy led to UFC normalization in 79 % of patients with no differences between the groups; only one patient failed to respond at all. Neither drug succeeded in controlling the disease when taken alone. LNSC dropped when compared to baseline levels, but not to a significant degree (p = 0.06), and it remained significantly higher than in controls (p = 0.0006). Associating cabergoline with ketoconazole may represent an effective second-line treatment, achieving a satisfactory reduction in UFC levels and clinical improvement. Although the combined treatment lowered patients' LNSC levels, they remained higher than normal, indicating a persistent subclinical hypercortisolism; the implications of this condition need to be considered. No differences emerged between the two treatment schedules.