Article Summary
MRI and ultrasound provide different measurements of uterine fibroids, with MRI tending to underestimate subserosal fibroids while ultrasound tends to overestimate them. Accurate preoperative imaging assessment is important for minimizing surgical complications during fibroid removal procedures.
- MRI underestimated while ultrasound overestimated the number and size of subserosal fibroids compared to actual surgical findings
- Study of 164 women found 42.1% used MRI and 40.9% used combined ultrasound and MRI for preoperative fibroid mapping
- Greater discrepancy between MRI estimates and actual fibroid size correlated with increased perioperative blood loss requiring cell salvage
- Neither imaging technique is clearly superior for suspected extrauterine fibroids; choice should consider insurance, surgical history, and other uterine pathologies
Associations between imaging and operative findings in uterine fibroids may differ by imaging modality, according to findings published July 24 in the Journal of Minimally Invasive Gynecology.
MRI underestimated while ultrasound overestimated the number and size of subserosal fibroids relative to intraoperative estimates in women undergoing laparoscopic-assisted robotic myomectomy, wrote a team led by Maeve McNamara, MD, from Emory University in Atlanta, GA.
“Our findings do not show that MRI or ultrasound are clearly superior when extrauterine fibroids are suspected, but reinforce that these types of fibroids, when encountered, are associated with significant intraoperative complications,” McNamara and colleagues wrote.
Robotic-assisted laparoscopic myomectomy may lead to longer operative times, but benefits include reduced complication rates relative to abdominal approaches in women seeking uterine-sparing surgery. Accurate pre-operative assessment of fibroid location and size is needed to minimize complications.
While ultrasound is typically used for preoperative fibroid evaluation, there is growing interest in using MRI for this purpose, including for distinguishing FIGO type 4 and 5 leiomyomas.
The McNamara team characterized the prevalence of ultrasound and MRI prior to robotic assisted-laparoscopic myomectomy. It also studied associations between pre-operative imaging and intraoperative estimates and correlated preoperative and intraoperative measurement agreements with clinical outcomes.
The study included 164 women with an average age of 37.4 years, with about 85% being non-Hispanic African American women.
The researchers reported the following findings:
The imaging methods most often used for preoperative fibroid mapping included MRI (42.1%) and combined ultrasound and MRI (40.9%).
MRI underestimated while ultrasound overestimated the number of submucosal fibroids relative to operative findings (ultrasound: M = -0.1; MRI: M = 0.9; p = 0.014).
MRI underestimated while ultrasound overestimated the size of subserosal fibroids relative to operative findings (Ultrasound: M = -1.4 cm; MRI: M= 0.4 cm; p = 0.008).
Cell salvage showed a statistically significant difference among post-operative complications that the researchers evaluated. The team found that greater discrepancy between MRI and operative estimates of subserosal fibroid size was tied to greater volume of perioperative cell salvage (low: M =103.1 mL; high: M = 187.5 mL; p = 0.046).
The study authors highlighted potential considerations for preoperative imaging. These include insurance coverage, prior surgical history, and suspicion for other uterine pathologies.
"While… guidelines propose a stepwise approach for preoperative imaging, utilizing ultrasound for initial screening and MRI for further characterization of fibroids for surgical candidates, our findings suggest there may a role for exclusive ultrasound in some surgical candidates,” the authors wrote. “The real-time, dynamic nature of ultrasound may better facilitate visualization of small subserosal and submucosal fibroids given that MRI tended to underestimate the size of subserosal fibroids and number of submucosal fibroids.”
Read the full study here.
![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&fit=crop&h=100&q=70&w=100)






![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&fit=crop&h=112&q=70&w=112)









