Radiology Obstetric Imaging

[Obstetric Imaging] MR Lung Volume in Fetal Congenital Diaphragmatic Hernia: Logistic Regression Analysis--Mortality and Extracorporeal Membrane Oxygenation

Busing, K. A., Kilian, A. K., Schaible, T., Dinter, D. J., Neff, K. W. — 2008-06-19

Purpose: To prospectively assess the results of logistic regression analysis that were based on magnetic resonance (MR) image fetal lung volume (FLV) measurements to predict survival and the corresponding need for extracorporeal membrane oxygenation (ECMO) therapy in fetuses with congenital diaphragmatic hernia (CDH) before and after 30 weeks gestation.

Materials and Methods: Written informed consent was obtained and the study was approved by the local research ethics committee. FLV was measured on MR images in 95 fetuses (52 female neonates, 43 male neonates) with CDH between 22 and 39 weeks gestation by using multiplanar T2-weighted half-Fourier acquired single-shot turbo spin-echo MR imaging. On the basis of logistic regression analysis results, mortality and the need for ECMO therapy were calculated for fetuses before and after 30 weeks gestation.

Results: Overall, higher FLV was associated with improved survival (P < .001) and decreasing probability of need for ECMO therapy (P = .008). Survival at discharge was 29.2% in neonates with an FLV of 5 mL, compared with 99.7% in neonates with an FLV of 25 mL. The corresponding need for ECMO therapy was 56.1% in fetuses with an FLV of 5 mL and 8.7% in fetuses with an FLV of 40 mL. Prognostic power was considerably lower before 30 weeks gestation.

Conclusion: Beyond 30 weeks gestation, logistic regression analysis that is based on MR FLV measurements is useful to estimate neonatal survival rates and ECMO requirements. Prior to 30 weeks gestation, the method is not reliable and the FLV measurement should be repeated, particularly in fetuses with small lung volumes, before a decision is made about therapeutic options.

[Obstetric Imaging] MR Relative Fetal Lung Volume in Congenital Diaphragmatic Hernia: Survival and Need for Extracorporeal Membrane Oxygenation

2008-06-19

Purpose: To retrospectively evaluate the accuracy of the absolute fetal lung volume (FLV) measured at magnetic resonance (MR) imaging and seven formulas for calculating relative FLV to predict neonatal survival and the need for extracorporeal membrane oxygenation (ECMO) in fetuses with congenital diaphragmatic hernia (CDH).

Materials and Methods: This retrospective study was approved by the research ethics committee, and informed consent was received from all mothers for previous prospective studies. In total, 68 fetuses with CDH were assessed by using MR image FLV measurement within 23–39 weeks gestation. The relative FLV was expressed as a percentage of the predicted lung volume calculated with biometric parameters according to seven formulas previously described in the literature. Applying the area under the curve (AUC), the various relative FLVs and the absolute FLV were investigated for their prognostic accuracy to predict neonatal survival and the need for ECMO therapy.

Results: All relative FLVs and the absolute FLV revealed a significant difference in mean lung volume between neonates who survived and neonates who did not survive (P = .001 to P < .001) and measurement accuracy was excellent for each method (AUC, 0.800–0.900). For predicting neonatal ECMO requirement, differences in FLVs were smaller but still significant (P = .05 to <.009) and measurement accuracy was acceptable throughout (AUC, 0.653–0.739).

Conclusion: The various relative FLVs and the absolute FLV measured at MR planimetry are each highly valuable in predicting survival in fetuses with CDH. For predicting whether neonatal ECMO therapy is required, the accuracy of the absolute FLV (AUC, 0.68) and that of the relative FLVs (AUC, 0.653–0.739) was acceptable.

[Obstetric Imaging] Frequency and Cause of Disagreements in Diagnoses for Fetuses Referred for Ventriculomegaly

2008-04-22

Purpose: To prospectively assess the frequency and cause of disagreements in diagnoses at ultrasonography (US) and magnetic resonance (MR) imaging for fetuses referred for ventriculomegaly (VM).

Materials and Methods: One hundred ninety-five women, aged 18–44 years, with 200 fetal referrals for VM, were recruited in a prospective IRB-approved, HIPAA-compliant study. Written informed consent was obtained. US scans were prospectively interpreted by three obstetric radiologists and MR examinations were read by one obstetric radiologist and three pediatric neuroradiologists. Final diagnosis was reached by consensus (198 US, 198 MR, and 196 US-MR comparisons). Gestational age, ventricular size, types of disagreements, and reasons for disagreements were recorded. Interreader agreement was assessed with statistics. Ventricular diameter, gestational age, and confidence scores were analyzed by using mixed-model analysis of variance, accounting for correlation within reader and fetus.

Results: There was prospective agreement on 118 (60%) of 198 US and 104 (53%) of 198 MR readings. Consensus was more likely when the final diagnosis was isolated VM (83 of 104, 80% at US; 82 of 109, 75% at MR) than when the final diagnosis included other anomalies as well (14 of 63, 22% at US; seven of 68, 10% at MR; P < .001). There was disagreement on 19 (10%) of 196 and 31 (16%) of 196 fetuses about the presence of VM at US and MR, respectively, and on 29 (15%) of 198 and 39 (20%) of 198 fetuses regarding the presence of major findings at US and MR, respectively. Reasons for discrepancies in reporting major findings included errors of observation, lack of real-time US scanning, lack of neuroradiology experience, as well as modality differences in helping depict abnormalities.

Conclusion: Of radiologists who read high-risk obstetric US and fetal MR images for VM, there is considerable variability in central nervous system diagnosis.

[Obstetric Imaging] Fetal Body Volume at MR Imaging to Quantify Total Fetal Lung Volume: Normal Ranges

2008-03-27

Purpose: To prospectively determine normal ranges of total fetal lung volume (TFLV) based on fetal body volume (FBV) and to determine whether prediction of TFLV based on such ranges is independent of fetal biometric indexes.

Materials and Methods: The study was approved by the Ethics Committee on Clinical Studies; informed consent was obtained. Magnetic resonance imaging volumetric measurement of fetal lung, liver, and body was performed in 200 fetuses without abnormalities affecting these structures. FBV was assessed with planimetric measurements by using T2-weighted half-Fourier rapid acquisition with relaxation enhancement at 16–40 weeks of gestation. TFLV was correlated to gestational age (GA), liver volume, and FBV. Observed-expected (O/E) ratio for TFLV was calculated by expressing the observed TFLV as a percentage of the expected mean TFLV for GA, liver volume, or FBV. Three groups of fetuses were defined on the basis of biometric percentiles for fetal weight obtained through ultrasonography: fetuses with weight at or below the 5th percentile, those with weight at or above the 95th percentile, and those with weight between these two percentiles (eutrophic). Median O/E ratios, based on GA and FBV, in fetuses with weight below the 5th percentile and in those with weight above the 95th percentile, were compared with median O/E ratio of eutrophic fetuses (Mann-Whitney U test).

Results: TFLV correlated best with FBV, according to the following cubic fit: TFLV = [(2.0 · 10^–9) · FBV³] – [(1.19 · 10^–5) · FBV²] + (0.0508 · FBV) – 1.79 (r² = 0.85, P < .001). In 174 eutrophic fetuses, normal median O/E ratio based on GA was 99.1% (range, 31.2%–158.0%), which was higher than that in 11 fetuses with weight at or below the 5th percentile (46.2%; range, 15.7%–87.3%) (P < .01) and lower than that in 15 fetuses with weight at or above the 95th percentile (146.8%; range, 87.2%–204.2%) (P < .01). Normal median O/E ratio, based on FBV, was independent of biometric indexes irrespective of the percentile for fetal weight.

Conclusion: FBV correlated best with TFLV, irrespective of biometric variables.