2054. Findings on T1 MR Imaging Associated With Cesarean Hysterectomy in Patients With Suspected Placenta Accreta Spectrum (PAS)
Authors* Denotes Presenting Author
Abiola Femi-Abodunde *;
University of Texas Southwestern Medical Center
Quyen Do;
University of Texas Southwestern Medical Center
Yin Xi;
University of Texas Southwestern Medical Center
Christina Herrera;
University of Texas Southwestern Medical Center
Catherine Spong;
University of Texas Southwestern Medical Center
Diane Twickler;
University of Texas Southwestern Medical Center
Objective:
T1-weighted imaging is an MR acquisition employed to evaluate certain characteristics of pathology, most commonly subacute blood, fat and proteinaceous fluid. T1 imaging is also particularly sensitive to the susceptibility of metal, which can lead to imaging artifacts. We have antidotally observed intra-placental bleeds on imaging via increased T1 intra-placental bright signal and pathology in the setting of placenta accreta spectrum (PAS). At our institution, we routinely employ T1-weighted sequences in our fetal MRI protocols. Our purpose is to describe T1-weighted MRI findings of the human placenta in the setting of placenta accreta spectrum (PAS) and assess the relationship to invasion requiring hysterectomy.
Materials and Methods:
MR studies with sagittal breath-hold T1-weighted images performed for suspicion of PAS were assessed by the following parameters: presence or absence of a skin scar, presence or absence of bright signal in the cervix , degree of intra-placental bright signal (0 - 3), sagittal length of the placenta in proximity to potential uterine scar (mm), degree of susceptibility artifact from previous cesarean deliveries (0 - 3), and presence or absence of a bulge, based on deformity of the uterine contour. Statistical analysis included two sample t-test, Wilcoxon rank sum test, exact Chi squared and Fisher exact test.
Results:
Of 86 women with available imaging correlated to clinical and pathologic outcomes, 57 had invasion resulting in hysterectomy, and 29 did not. Significant variables included severity of artifact grading (p = 0.02), and presence of uterine contour distortion/bulge (p = 0.02). The sagittal length of the placenta in proximity to the suspected uterine scar was greater in those requiring hysterectomy, but did not reach statistical significance, p = 0.05, mean 78.7 +/ -46.6 (hysterectomy) vs 62.0 +/ -44.3 (no hysterectomy). The skin scar, bright signal in the cervix and intra-placental bright signal were not significantly different.
Conclusion:
The sagittal MR T1-weighted breath hold sequence provided important findings in women with previous cesarean delivery at risk for PAS resulting in hysterectomy by demonstrating and grading the artifacts of the previous surgeries. Although infrequent, presence of the bulge or contour abnormality is also significant. Proximity of the placenta to the scar artifacts approached significance in this small series. We recommend routine T1-weighted sagittal MRI to serve as a roadmap, used in conjunction with the standard half-Fourier RARE (rapid acquisition with relaxation enhancement) T2 weighted image for recommendations to the multidisciplinary PAS care team.
