First-trimester choroid-plexus-to-lateral-ventricle disproportion and prediction of subsequent ventriculomegaly.

Abstract

Ventriculomegaly can be associated with long-term neurodevelopmental impairment. Prenatal diagnosis of ventriculomegaly is most commonly made at the routine second-trimester anomaly scan. The value of first-trimester ultrasound has expanded to early diagnosis and screening of fetal abnormalities. The objective of this study was to assess the predictive accuracy of first-trimester choroid-plexus-to-lateral-ventricle-or-head ratios for development of ventriculomegaly at a later gestational age. This was a case-control study of fetuses with isolated ventriculomegaly diagnosed after 16 weeks' gestation and a control group of normal fetuses (without ventriculomegaly). The exclusion criteria included aneuploidy, genetic syndrome and/or other brain abnormality. Stored two-dimensional first-trimester ultrasound images were analyzed blindly offline and fetal biometry was performed in the axial view of the fetal head. The ratios of choroid plexus area (PA) to lateral ventricular area (VA), choroid plexus length (PL) to lateral ventricular length (VL), choroid plexus diameter (PD) to lateral ventricular diameter (VD) and PA to biparietal diameter (BPD) were measured at 11 + 0 to 13 + 6 weeks' gestation. Intra- and interobserver variability of measurement of these fetal head biometric parameters at 11 + 0 to 13 + 6 weeks' gestation were assessed in 20 normal fetuses using intraclass correlation coefficients with 95% CI. The accuracy of first-trimester biometric measurements for prediction of ventriculomegaly was assessed using the area under the receiver-operating-characteristics curves (AUC). The analysis included 683 singleton pregnancies, of which 102 fetuses were diagnosed with ventriculomegaly. Ventriculomegaly was mild in 86 (84.3%) cases and severe in the other 16 (15.7%). All first-trimester fetal choroid-plexus-to-lateral-ventricle/head ratios were significantly lower in cases with ventriculomegaly compared with controls (P < 0.001), with good inter- and intraobserver agreement (≥ 0.95) for the majority of the fetal head biometric parameters assessed. On adjusting for crown-rump length, optimism-adjusted AUC values obtained after cross-validation showed that both PL/VL ratio (AUC, 0.87 (95% CI, 0.73-0.98)) and PA/VA ratio (AUC, 0.90 (95% CI, 0.82-0.98)) had good predictive accuracy for severe ventriculomegaly. The PA/BPD ratio (AUC, 0.73 (95% CI, 0.54-0.90)) had modest predictive ability, which was significantly lower compared with that of the PA/VA ratio and PL/VL ratio (P = 0.003 and P = 0.001, respectively). The predictive accuracy of PD/VD ratio was low with an AUC of 0.65 (95% CI, 0.47-0.84). Optimism-adjusted AUC values obtained after cross-validation showed that PA/VA ratio offered the highest predictive accuracy for mild ventriculomegaly with an AUC of 0.84 (95% CI, 0.79-0.89), followed by PL/VL ratio (AUC, 0.82 (95% CI, 0.76-0.88)), PA/BPD ratio (AUC, 0.76 (95% CI, 0.69-0.82)) and PD/VD ratio (AUC, 0.75 (95% CI, 0.67-0.81)). Calibration plots showed that both PA/VA and PL/VL ratios had good calibration. First-trimester prediction of ventriculomegaly using ratios of fetal choroid plexus to lateral ventricle/head appears promising. Future prospective studies are needed to validate the predictive accuracy of these ultrasound markers as a screening tool for ventriculomegaly. © 2023 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.