Background: The utility of a repeated EUS by experts is not known.
Objective: To define the utility of a repeated EUS for the same indication.
Design: A retrospective case series.
Setting: Tertiary-referral hospital in Indianapolis, Indiana.
Patients: Consecutive subjects, with and without cancer, who, between January 2000 and September 2006, underwent an initial EUS elsewhere within 6 and 12 weeks of a repeated EUS at our hospital.
Interventions: A repeated EUS.
Main outcome measurements: Clinical impact of a repeated EUS.
Results: Of 8936 EUS examinations, 73 repeated procedures (0.8%) were identified, and 24 were excluded. The 49 initial EUS procedures (26 men, median age 59 years) were done in Indiana (n = 44) or another state (n = 5) by one of 15 physicians in private practice (n = 48) or at a teaching hospital (n = 1). An EUS-guided FNA (EUS-FNA) was performed during an initial EUS in 21 patients (no biopsy diagnostic for cancer) and was not attempted in 14 patients. The principle indication for a repeated EUS (n = 35) was for an EUS-FNA after the initial tissue sampling was benign, nondiagnostic, or not done. A second EUS had no clinical impact in 18 patients (37%). In the remaining 31 patients (63%), a repeated EUS provided a new or changed clinical diagnosis (n = 12), the initial diagnosis of primary pancreatic cancer (n = 5) or GI stromal tumor (GIST) (n = 1) after a previous nondiagnostic biopsy; or the initial diagnosis of primary (n = 4) or metastatic (n = 2) pancreatic cancer, metastatic esophageal cancer (n = 1), hilar cholangiocarcinoma (n = 1), GIST (n = 1), or pancreatic neuroendocrine tumor (n = 1), or an initial aspiration of a pancreatic cyst (n = 3) after a previous EUS-FNA was not able to be performed.
Limitations: A retrospective design; a small number of nonpancreatic indications.
Conclusions: In this study, a repeated EUS at a tertiary-referral center had a clinical impact in 63% of patients when performed by experts for a similar clinical indication.