Ultrasound versus Magnetic Resonance Imaging for Uterine Myoma Mapping before Robotic-assisted Laparoscopic...

To (1) characterize the prevalence of ultrasound (US) and magnetic resonance imaging (MRI) before robotic-assisted laparoscopic myomectomy; (2) investigate the association between preoperative imaging (US vs MRI) and intraoperative estimates of myoma size, number, and number multiplied by size (metric estimating myoma burden); and (3) correlate preoperative and intraoperative measurement agreement with clinical outcomes.

A retrospective chart review was conducted for patients who underwent robotic-assisted laparoscopic myomectomy for symptomatic uterine myomas at a single institution, by a single surgeon, from 2013 to 2023.

Large southeastern urban academic hospital.

164 patients (mean age = 37.4 ± 4.9; 85.4% non-Hispanic African American).

US or MRI before robotic-assisted laparoscopic myomectomy.

Individual myoma type, size, and number were defined according to radiographic and operative reports. Intraoperative and postoperative outcomes were abstracted. Differences were assessed between preoperative imaging and operative estimates of myoma number, size, and burden of all myomas in the cohort; results were stratified by anatomic type of myoma. Combined US/MRI (n = 67, 40.9%) and MRI only (n = 69, 42.1%) were used for preoperative myoma mapping by imaging most frequently. MRI tended to underestimate and US tended to overestimate the number of submucosal myomas relative to operative findings (US, mean = -0.1, standard deviation [SD] = 0.5; MRI, mean = 0.9, SD = 1.3; p = .014). In addition, MRI underestimated and US overestimated the size (in cm) of subserosal myomas relative to operative findings (US, mean = -1.4, SD = 3.0; MRI, mean = 0.4, SD = 3.0; p = .008). Of the postoperative complications evaluated, only cell salvage had a statistically significant difference; a greater discrepancy between MRI and operative estimates of subserosal myoma size was associated with greater volume of perioperative cell salvage (low, mean = 103.1, SD = 169.2; high, mean = 187.5, SD = 224.2; p = .046).

Among patients undergoing laparoscopic-assisted robotic myomectomy, this pilot study observed that MRI tended to underestimate and US tended to overestimate the number of submucosal myomas and the size of subserosal myomas relative to intraoperative estimates.