Comparison of diagnostic performance for perinatal and paediatric post-mortem imaging: CT versus MRI.

Abstract

To compare the diagnostic yield of whole-body post-mortem computed tomography (PMCT) imaging to post-mortem magnetic resonance (PMMR) imaging in a prospective study of fetuses and children. We compared PMCT and PMMR to conventional autopsy as the gold standard for the detection of (a) major pathological abnormalities related to the cause of death and (b) all diagnostic findings in five different body organ systems. Eighty two cases (53 fetuses and 29 children) underwent PMCT and PMMR prior to autopsy, at which 55 major abnormalities were identified. Significantly more PMCT than PMMR examinations were non-diagnostic (18/82 vs. 4/82; 21.9% vs. 4.9%, diff 17.1% (95% CI 6.7, 27.6; p < 0.05)). PMMR gave an accurate diagnosis in 24/55 (43.64%; 95% CI 31.37, 56.73%) compared to 18/55 PMCT (32.73%; 95% CI 21.81, 45.90). PMCT was particularly poor in fetuses <24weeks, with 28.6% (8.1, 46.4%) more non-diagnostic scans. Where both PMCT and PMMR were diagnostic, PMMR gave slightly higher diagnostic accuracy than PMCT (62.8% vs. 59.4%). Unenhanced PMCT has limited value in detection of major pathology primarily because of poor-quality, non-diagnostic fetal images. On this basis, PMMR should be the modality of choice for non-invasive PM imaging in fetuses and children. • Overall 17.1% more PMCT examinations than PMMR were non-diagnostic • 28.6% more PMCT were non-diagnostic than PMMR in fetuses <24weeks • PMMR detected almost a third more pathological abnormalities than PMCT • PMMR gave slightly higher diagnostic accuracy when both were diagnostic.