
When radiologists find incidental pancreatic cysts on MRI scans, patients can benefit from prompt treatment. But what happens when ordering physicians don't heed the advice of radiologists for follow-up? One outcome is that downstream costs can vary widely, according to an October 9 study in the American Journal of Roentgenology.
Among 200 patients, the average cost to follow-up incidental cysts ranged from $460 to $872, with the highest value due primarily to additional tests. When ordering physicians "overmanaged" these cases compared with radiologist recommendations, costs were on the higher side, at an average of $842 per incidental cyst. When physicians followed radiologists' advice, the cost decreased by 25%. Additionally, when they "undermanaged" patients based on American College of Radiology (ACR) incidental finding guidelines, cost dropped by more than 70%.
"The findings suggest a role for targeted educational efforts, collaborative partnerships, and other initiatives to foster greater adherence to radiologist recommendations, including critical test results notification systems, automated reminders within electronic health systems, and stronger language within radiology reports when no follow-up testing is recommended," wrote lead author Dr. Andrew Rosenkrantz and colleagues from NYU Langone Medical Center in New York City and Hackensack Radiology Group in River Edge, NJ.
Pancreatic cysts are seen in approximately 20% of abdominal MRI scans, according to previous research. The challenge is determining an accurate diagnosis and the need for follow-up treatment based on a cyst's characteristics at the time of initial detection. Therefore, follow-up CT, MRI, or endoscopic ultrasound scans may be needed to better evaluate the situation.
"The costs associated with such events are important to consider given the increasing and unsustainable expenditures of the U.S. healthcare system, the wide variability of care delivery, and the increasing federal mandate to focus on value in healthcare delivery to improve the nation's healthcare system," the authors noted.
The researchers began their study in January 2013 by reviewing the records of patients with incidental pancreatic cysts found on initial MRI scans. They investigated whether a radiologist recommended follow-up testing, such as additional imaging or endoscopic ultrasound. They then calculated the mean cost per cyst and how follow-up and severity of the case, such as a malignancy, affected the mean cost. Dollar amounts were estimated using national Medicare rates and a 3% annual discount rate.
Total medical costs for the 200 patients were $92,182, or $460 per cyst. Additional costs were greater when the ordering physician overmanaged a patient's care relative to the radiologist's recommendation, while costs fell when recommendations were followed.
| Cost variation for following up incidental pancreatic cysts on MRI scans | ||
| Ordering physician action | No. of patients | Average cost per cyst |
| As compared with radiologists' recommendations | ||
| Overmanagement | 24 | $842 |
| Adherence | 72 | $631 |
| Undermanagement | 102 | $252 |
| As compared with ACR incidental finding guidelines | ||
| Overmanagement | 36 | $845 |
| Adherence | 45 | $811 |
| Undermanagement | 119 | $212 |
"The implications of our work may extend beyond simply pancreatic cyst management itself by providing a research design that can be similarly applied to other common incidental imaging findings, including thyroid and lung nodules," Rosenkrantz and colleagues wrote. "We hope to catalyze a shift in thought process regarding such findings, whereby future investigators will also consider the economic effect and, hence, value of lesion follow-up when evaluating management algorithms for incidental findings."



![Overview of the study design. (A) The fully automated deep learning framework was developed to estimate body composition (BC) (defined as subcutaneous adipose tissue [SAT] in liters; visceral adipose tissue [VAT] in liters; skeletal muscle [SM] in liters; SM fat fraction [SMFF] as a percentage; and intramuscular adipose tissue [IMAT] in deciliters) from MRI. The fully automated framework comprised one model (model 1) to quantify different BC measures (SAT, VAT, SM, SMFF, and IMAT) as three-dimensional (3D) measures from whole-body MRI scans. The second model (model 2) was trained to identify standardized anatomic landmarks along the craniocaudal body axis (z coordinate field), which allowed for subdividing the whole-body measures into different subregions typically examined on clinical routine MRI scans (chest, abdomen, and pelvis). (B) BC was quantified from whole-body MRI in over 66,000 individuals from two large population-based cohort studies, the UK Biobank (UKB) (36,317 individuals) and the German National Cohort (NAKO) (30,291 individuals). Bar graphs show age distribution by sex and cohort. BMI = body mass index. (C) After the performance assessment of the fully automated framework, the change in BC measures, distributions, and profiles across age decades were investigated. Age-, sex-, and height-adjusted body composition reference curves were calculated and made publicly available in a web-based z-score calculator (https://circ-ml.github.io).](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/05/body-comp.XgAjTfPj1W.jpg?auto=format%2Ccompress&fit=crop&h=100&q=70&w=100)






![Overview of the study design. (A) The fully automated deep learning framework was developed to estimate body composition (BC) (defined as subcutaneous adipose tissue [SAT] in liters; visceral adipose tissue [VAT] in liters; skeletal muscle [SM] in liters; SM fat fraction [SMFF] as a percentage; and intramuscular adipose tissue [IMAT] in deciliters) from MRI. The fully automated framework comprised one model (model 1) to quantify different BC measures (SAT, VAT, SM, SMFF, and IMAT) as three-dimensional (3D) measures from whole-body MRI scans. The second model (model 2) was trained to identify standardized anatomic landmarks along the craniocaudal body axis (z coordinate field), which allowed for subdividing the whole-body measures into different subregions typically examined on clinical routine MRI scans (chest, abdomen, and pelvis). (B) BC was quantified from whole-body MRI in over 66,000 individuals from two large population-based cohort studies, the UK Biobank (UKB) (36,317 individuals) and the German National Cohort (NAKO) (30,291 individuals). Bar graphs show age distribution by sex and cohort. BMI = body mass index. (C) After the performance assessment of the fully automated framework, the change in BC measures, distributions, and profiles across age decades were investigated. Age-, sex-, and height-adjusted body composition reference curves were calculated and made publicly available in a web-based z-score calculator (https://circ-ml.github.io).](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/05/body-comp.XgAjTfPj1W.jpg?auto=format%2Ccompress&fit=crop&h=112&q=70&w=112)








