C. Frye1, Z. Song1, J. Osborne2, C. McLeod1, S. Tridandipani3, A. Gillis1, J. Fazendin1, H. Chen1, B. Lindeman1 1University Of Alabama at Birmingham, Department Of Surgery, Birmingham, Alabama, USA 2University Of Alabama at Birmingham, Department Of Computer Science, Birmingham, Alabama, USA 3University Of Alabama at Birmingham, Department Of Radiology, Birmingham, Alabama, USA
Introduction:
Adrenal gland incidentalomas (AGIs) are commonly found on imaging but rates of indicated workup are notoriously low. While published guidelines summarize the suggested workup, AGI management can be challenging, especially for non-specialists. Prior studies have suggested that the setting in which the AGI is detected (i.e. outpatient vs inpatient vs emergency room [ER]) could influence the rate of indicated workup. We hypothesized that outpatient-detected patients would have the highest rates of indicated workup and that ER-detected patients the lowest.
Methods:
Radiology reports from all relevant cross-sectional images performed between June 2020 and July 2021 were retrospectively analyzed by a natural language processing (NLP) algorithm designed to identify clinically significant AGIs. Clinical data from these patients were collected from June to September 2022. Statistical analyses were performed using chi-square, ANOVA, Kruskal-Wallis rank sum, and multivariable logistic regression (adjusted for sex, age, nodule size, and scan indication).
Results:
Of 50,529 radiology reports searched by the NLP algorithm, 1,784 (3.5%) patients were identified to have an AGI. Of these, the setting in which the AGI was discovered included outpatient (1,002; 56.2%), inpatient (526, 29.5%), and ER (256, 14.3%). Among the outpatient, inpatient, and ER settings, respectively, there were significant differences (all p<0.001) in sex (male 48%, 56%, and 42%), age (65.2, 64.3, and 59.5 years), nodule size (1.8, 1.9, and 1.6 cm), biochemical workup (19%, 6%, and 9%), adrenal-specific follow-up imaging (18%, 2%, and 6%), and any non-adrenal follow-up imaging (57%, 48%, 44%). A significant difference was also observed in the scan indications (p<0.001), with the most common indication for each setting being: malignancy, outpatient (44%); medical conditions/miscellaneous, inpatient (45%); and pain, ER (63%). With outpatient as reference, multivariable logistic regression revealed that biochemical workup was significantly less likely to occur for inpatient-discovered (OR 0.44, CI 0.26-0.72, p<0.01) but not ER-discovered patients (OR 0.71, CI 0.37-1.31, p=0.28). Adrenal-specific follow-up scans were significantly less likely to occur in both inpatient-detected (OR 0.18, CI 0.08-0.34, p<0.001) and ER-detected (OR 0.40, CI 0.19-0.83, p=0.02) patients. ER-discovered patients were also significantly less likely to undergo any non-adrenal follow-up imaging (OR 0.58, CI 0.39-0.86, p<0.01).
Conclusion:
The setting in which an AGI is detected appears to have implications for further workup rates. Specifically, AGIs discovered in the inpatient setting were the least likely to receive guideline-adherent workup (biochemical workup and adrenal-specific imaging), followed by those found in the ER (adrenal-specific imaging only). Patients who have AGIs discovered in the inpatient and ER settings should receive extra attention to avoid inadequate evaluation.