Maximizing the accuracy of Inferior petrosal sinus sampling: validation of the use of Prolactin as a marker of pituitary venous effluent in the diagnosis of Cushing's disease.

OBJECTIVE: This study aimed to increase the accuracy of the inferior petrosal sinus sampling (IPSS) procedure and reduce the false-negative rate through the addition of prolactin as a marker of pituitary venous outflow as well as validate this adjunct to the test process.

CONTEXT: Inferior petrosal sinus sampling (IPSS) for ACTH is the current gold standard test for the differentiation of pituitary Cushing's disease from the ectopic ACTH syndrome. Although early studies with IPSS reported a diagnostic sensitivity and specificity approaching 100%, additional experience has revealed a false-negative rate of 1-10%. This has been attributed to either technical problems with unsuccessful petrosal sinus catheterization or anomalous venous drainage of the pituitary. Previous studies have suggested that the measurement of other anterior pituitary hormones may be useful during IPSS as a guide to the effectiveness of cannulation and to improve the diagnostic accuracy of the procedure.

DESIGN: We reviewed the data, in this retrospective cohort study, for all patients who had undergone IPSS for the investigation of ACTH-dependent hypercortisolism.

PATIENTS: The study included 83 patients who underwent IPSS at St. Thomas's hospital between 2005 and 2010.

MEASUREMENTS: Plasma ACTH and prolactin levels were measured both centrally and peripherally. The normalized ACTH/Prolactin inferior petrosal sinus/peripheral ratio was then calculated to assess the accuracy of the sampling procedure.

RESULTS: A total of 83 patients with confirmed ACTH-dependent cortisol excess underwent investigation with IPSS during the study period. Sixty-seven patients initially had a positive IPSS result (i.e. a basal central/peripheral ACTH ratio >2·0 and >3·0 post-CRH). However, when the concurrent prolactin data were used, six patients were additionally found to have positive results suggestive of pituitary Cushing's. The Prolactin normalized ACTH IPS/Peripheral ratios were all >0·8 in patients with proven Cushing's disease, whereas they were all <0·6 in proven ectopic ACTH syndrome. The diagnosis was subsequently confirmed histologically in 72 of the patients.

CONCLUSIONS: Using Prolactin as a concurrent index of pituitary venous effluent helps us recognize whether pituitary venous blood has been accurately sampled. Normalizing the IPS/peripheral ratios with Prolactin helps to improve the accuracy of the result and reduces the false-negative rate. With regards to the usefulness/validity of this test in clinical practice, it is relevant, reproducible and is easily adaptable from the existing diagnostic sequence.