New England Surgical Society (NESS)
Search NESS
  Annual Meeting
  Annual Resident and Fellow Research Day
      Member Directory
      Members Only
  Scholars Foundation
  Job Board
  Journal of the American College of Surgeons
  Contact NESS

2010 Annual Meeting Abstracts

Back to Program

Preoperative adrenal imaging: Can we rely on it when approaching adrenal tumors?
*Peter S Wu1, *Jane L Garb1, Richard B Wait1, *John I Lew2, *Carmen C Solorzano2, *Azad A Jabiev1
1Baystate Medical Center, Springfield, MA;2University of Miami Miller School of Medicine, Miami, FL

Objective: The approach to nonfunctional adrenal tumors is based largely on size as determined by CT or MRI. The objective of this study was to determine the accuracy of preoperative adrenal imaging. Design: Retrospective review. Setting: University and university-affiliated tertiary hospitals. Patients: Retrospective review of 189 patients who underwent adrenalectomy from 2000 to 2008 was conducted. From 192 identified tumors, 185 were eligible for analysis. Main outcome measures: Spearman coefficient (r) was used to gauge the degree of correlation between imaged and true tumor size using the largest dimension available from radiologic and pathologic reports. Wilcoxon matched pairs test was performed to determine the difference between imaged and actual tumor sizes. Using multiple regression analysis, the effect of tumor size on accuracy of CT imaging was determined for tumors ≤6cm and >6cm. A p value <0.05 was considered statistically significant. Results: There was a strong correlation between CT image and actual tumor sizes (n=154, r=0.72, 95%CI 0.63-0.79, p<0.01) although CT measurements significantly underestimated actual sizes (4.33±2.56 vs. 4.67±2.95cm, p<0.05). Further analysis demonstrated that tumors ≤6cm on CT were underestimated by an average of 0.46cm while tumors >6cm were overestimated by 0.33cm (p<0.01). Similarly MRI size strongly correlated with actual tumor size (n=80, r=0.71, 95%CI 0.57-0.80, p<0.01) however MRI also underestimated true size (3.89±2.47 vs. 4.20±2.59cm, p<0.01). For both imaging modalities, 52.4% of all tumors were underestimated in size. Furthermore 42.7% of tumors were undersized by ≥0.5cm and 27.6% were undersized by ≥1.0cm. Conclusions: Preoperative adrenal imaging may not accurately predict actual tumor size. CT and MRI underestimate the sizes of tumors under 6cm. These findings should be considered when managing adrenal tumors non-operatively.

Back to Program


Copyright © 2020 New England Surgical Society. All Rights Reserved.
Read Privacy Policy.