Prolactin may reduce false-negative results in diagnosing Cushing's disease (CD) during inferior petrosal sinus sampling (IPSS). Prolactin normalization could improve the accuracy of IPSS in predicting adenoma lateralization in CD. However, none of the previous studies had involved the use of desmopressin during IPSS. Our objective was to examine the utility of prolactin measurement during IPSS with desmopressin stimulation. We conducted a retrospective analysis of 40 patients (including 31 females) with ACTH-dependent Cushing's syndrome who underwent IPSS between 2010 and 2013. Thirty-eight CD patients were partitioned into true positive (n = 35) and false negative (n = 3). The proportion of improper IPSS venous sampling defined by corresponding IPS:P (inferior petrosal sinus to peripheral) prolactin ratio <1.8 was significantly different between two groups (P = 0.004). Applying a prolactin-normalized ACTH IPS:P ratio >0.8 cutoff could increase the sensitivity of IPSS to 38/38 (100 %). Among the 31 patients with histopathologically proven adenoma localization, correct prediction of adenoma lateralization was obtained in 14/31 (45 %) patients by a peak intersinus ACTH gradient of ≥1.4 in baseline and was not improved by desmopressin stimulation. Left-right intersinus gradients of unilateral prolactin-adjusted ACTH IPS:P ratios could increase the correct prediction of adenoma lateralization to 20/31 (65 %) in baseline and 24/31 (77 %) (P = 0.006) after desmopressin stimulation, respectively. Prolactin is helpful to adjust negative results of IPSS with desmopressin stimulation. It may improve the accuracy in predicting adenoma lateralization in CD as well.