S. Z. Wennmacker1, E. De Savornin Lohman1, P. De Reuver1, N. Hasami1, M. Boermeester2, J. Verheij3, E. Spillenaar Bilgen4, J. Meijer5, K. Bosscha6, H. Van Der Linden7, I. Nagtegaal8, J. Drenth9, C. Van Laarhoven1 1Radboudumc,Surgery,Nijmegen, Netherlands 2AmsterdamUMC,Surgery,Amsterdam, Netherlands 3AmsterdamUMC,Pathology,Amsterdam, Netherlands 4Rijnstate Hospital,Surgery,Arnhem, Netherlands 5Rijnstate Hospital,Pathology,Arnhem, Netherlands 6Jeroen Bosch Hospital,Surgery,’s-Hertogenbosch, Netherlands 7Jeroen Bosch Hospital,Pathology,’s-Hertogenbosch, Netherlands 8Radboudumc,Pathology,Nijmegen, Netherlands 9Radboudumc,Gastroenterology And Hepatology,Nijmegen, Netherlands
Introduction: Cholecystectomy is only needed for neoplastic gallbladder polyps, in order to halt or prevent the development of gallbladder cancer. Current international guidelines advocate surgery for all gallbladder polyps ≥1cm, in view of the elevated risk of neoplasia of these polyps. However, the validity of this threshold may be questioned as one third of the polyps are wrongly classified. The aim of this study was to identify preoperative clinical and imaging characteristics associated with neoplastic polyps. Secondly, the concordance between imaging findings and pathological findings of gallbladder polyps was assessed.
Methods: A retrospective analysis of all histopathologically proven gallbladder polyps in four Dutch hospitals between 2003-2013 was performed. Patients were identified through PALGA, the Dutch nationwide network and registry of histo- and cytopathology. Clinical and imaging characteristics of patients with neoplastic versus nonneoplastic polyps were assessed using univariable and multivariable analysis. Concordance of polyp size, number of polyps and polyp type on ultrasound and histopathology were assessed using McNemars’ test, and subsequent sensitivities and specificities were calculated.
Results: A total of 208 patients were included of whom 95 patients (43.7%) were diagnosed with a neoplastic polyp on histopathological evaluation. Patients’ age (OR 1.06 per year (1.03-1.08), p<0.001) and a history of gallbladder disease (OR 2.80 (1.03-7.62), p=0.04) were significantly associated with neoplastic polyps on multivariable analysis. A total of 156 patients underwent preoperative ultrasound. In 88 patients (56.4%) the polyps were preoperatively identified on ultrasound. Polyp characteristics (shape, surface, echogenicity and internal echogenic pattern) were frequently unreported in imaging reports. Polyp size and number of polyps on ultrasound were significantly associated with neoplastic polyps in univariable, but not in multivariable analysis. Polyp size as assessed on ultrasound was significantly inconsistent with size on histopathology (p=0.002). Sensitivity and specificity of ultrasound for polyp size ≥1cm were 96% and 41%, for presence of a single polyp 92% and 62%, and for identifying neoplastic polyp type 70% and 25%.
Conclusion: Patients >50 years of age and with a history of gallstone disease are more likely to have neoplastic polyps. Current standard preoperative (ultrasound) imaging has low specificity for polyp size, presence of a single polyp, and the overall diagnostic accuracy of ultrasound in establishing polyp type is poor, risking considerable surgical overtreatment of patients with nonneoplastic polyps.