O. MORENO-ALVAREZ*, E. HERNANDEZ-ANDRADE*, D. OROS*, J. JANI, J. DEPREST and E. GRATACOS*

Transcription

1 Ultrasound Obstet Gynecol 2008; 31: Published online 14 January 2008 in Wiley InterScience ( DOI: /uog.201 Association between intrapulmonary arterial parameters and degree of lung growth as measured by lung-to-head ratio in fetuses with congenital diaphragmatic hernia O. MORENO-ALVAREZ*, E. HERNANDEZ-ANDRADE*, D. OROS*, J. JANI, J. DEPREST and E. GRATACOS* *Fetal and Perinatal Medicine Research Group, Department of Maternal Fetal Medicine, Institut Clinic de Ginecologia, Obstetricia i Neonatologia (ICGON), Hospital Clínic, Institut d Investigacions Biomediques August Pi i Sunyer (IDIBAPS) Universitat de Barcelona, and Centre for Biomedical Research on Rare Diseases (CIBERER), Instituto de Salud Carlos III, Barcelona, Spain and Department of Obstetrics and Gynecology, University Hospital Gasthuisberg, Leuven, Belgium KEYWORDS: congenital diaphragmatic hernia; fetal lung; lung-to-head ratio; peak early-diastolic reversed flow; pulsed ABSTRACT Objectives To describe the association between the intrapulmonary circulation and the degree of lung growth in fetuses with congenital diaphragmatic hernia (CDH). Methods Thirty-six fetuses with isolated left CDH at weeks of gestation were studied. Four groups were defined according to lung-to-head ratio (LHR), expressed as the ratio of observed to expected LHR for gestational age (O/E LHR). Pulsatility index (PI), peak systolic velocity (PSV), peak early-diastolic reversed flow (PEDRF), end-diastolic velocity (EDV), acceleration time (AT), time averaged velocity (TAV) and the ratio AT to ejection time (AT/ET) were measured in all fetuses in the proximal arterial branch of both lungs. For analysis, values were converted to Z-scores. Results It was possible to record all parameters from the contralateral lung of all CDH fetuses, and from the ipsilateral lung in 28 (77.7%). PEDRF significantly increased in relation to the decrease in O/E LHR in both lungs (contralateral: r = 0.6, P = 0.01; ipsilateral: r = 0.43, P = 0.02). PI increased significantly with the reduction in O/E LHR but only on the side contralateral to the hernia (r = 0., P = 0.02). In the ipsilateral pulmonary artery, PI values showed a significant overall increase but no differences were observed among the O/E LHR groups. PSV, EDV and TAV were reduced in both lungs with respect to normal values, but no association was found with the O/E LHR. No changes in AT or AT/ET were observed in either lung. Conclusions PEDRF is quick and easy to calculate, and showed the strongest and most consistent correlation with O/E LHR. The association between PEDRF and LHR might be an additional parameter that could help to establish prognosis in fetuses with CDH. Copyright 2008 ISUOG. Published by John Wiley & Sons, Ltd. INTRODUCTION Isolated congenital diaphragmatic hernia (CDH) is associated with a neonatal mortality rate of about 60% mainly due to pulmonary hypoplasia and pulmonary hypertension 1 3. The early compression produced by the abdominal organs in the lung located on the side of the hernia (ipsilateral) and on the opposite (contralateral) side might affect normal development of the vascular and bronchial trees, either by a mechanical effect or by interfering with the biochemical process of maturation 4,. The main parameter used to evaluate the severity of CDH is the lung-to-head ratio (LHR) 6. Recently, the use of the LHR has been assessed by calculating the O/E LHR, which is the ratio of the LHR observed to the LHR expected for gestational age (GA). This new approach corrects for the effect of GA, allowing a more accurate estimation of pulmonary hypoplasia mainly in Correspondence to: Dr E. Hernandez-Andrade, Fetal and Perinatal Medicine Research Group, Department of Maternal Fetal Medicine, Hospital Clínic, University of Barcelona, Sabino de Arana 1, Edificio Helios 2, Barcelona, Spain ( EHERNANDEZ@clinic.ub.es; powerdoppler@hotmail.com) Accepted: July 2007 Copyright 2008 ISUOG. Published by John Wiley & Sons, Ltd. ORIGINAL PAPER

2 in congenital diaphragmatic hernia 16 the most affected cases (O/E LHR < 2%) 7. However, its predictive capacity at intermediate O/E LHR values is not very accurate, and there is a real possibility that fetuses with a similar O/E LHR might have different prognoses. Therefore, the O/E LHR should be complemented with alternative parameters to improve prognosis in fetuses with CDH. Based on the concept that normal lung development involves the simultaneous growth of airways and blood vessels 8, several authors have suggested that the intrapulmonary circulation is altered in fetuses that develop pulmonary hypoplasia However, to date, no studies have correlated parameters with well established predictors of mortality in CDH, such as the LHR. The aim of the present study was to establish reference values for parameters in the main branch of the intrapulmonary artery, to evaluate the feasibility of measuring these parameters in both lungs of fetuses with CDH, and to explore the potential correlation between parameters and lung size, as evaluated by the O/E LHR, in order to assess their potential use in establishing the prognosis of CDH. METHODS Thirty-six singleton pregnancies complicated with isolated left CDH but with a normal karyotype were studied. None of the fetuses showed other structural abnormalities. CDH was confirmed at autopsy or during surgery for correction of the defect. At diagnosis the median maternal age was 32 (range, 23 37) years and the median GA was 2 + (range, to29+ 2) weeks. To construct normal reference values for the parameters, the proximal arterial branches of the pulmonary arteries in 12 normally grown fetuses at 20 3 weeks of gestation were analyzed in the side with the best angle of insonation. The study was approved by the hospital ethics committee and written consent was obtained for study of all fetuses. Ultrasound and studies were performed at the time of diagnosis with Siemens Sonoline Antares (Siemens Medical Systems, Malvern, PA, USA) or Voluson 730 Expert (GE Medical Systems, Milwaukee, WI, USA) ultrasound equipment with 6 2- or 7 2-MHz curved probes. All examinations were done in the absence of fetal corporeal or respiratory movements and with the mother in voluntary suspended respiration. For the studies the angle of insonation was maintained below 30. The high-pass filter was set at 120 Hz or alternatively at 70 Hz, and the mechanical and thermal indices were maintained below 1. The lung contralateral to the CDH defect was visualized in a cross-sectional image of the fetal thorax in the plane of the cardiac four-chamber view. The LHR was estimated using the two major perpendicular planes of the lung as proposed by Metkus et al. 12. The expected (E) LHR for GA was calculated according to Jani et al. 7 in the right lung as GA ( GA 2 ), and in the left lung as GA ( GA 2 ). The O/E LHR was calculated as LHR observed/lhr expected 100. For the analysis, the fetuses were divided into four groups based on the O/E LHR and expected survival: Group 1, n = 3(O/ELHR< 1%, survival 0%); Group 2, n = 6 (O/E LHR 16 2%, survival 1%); Group 3, n = 12 (O/E LHR 26 3%, survival 6%); and Group 4, n = 1 (O/E LHR > 36%, survival > 80%) 7. The intrapulmonary circulation of both lungs was visualized by color directional or power ultrasound imaging. The contralateral lung was easily recognized in the side of the thorax in which the heart was located (Figure 1a). The ipsilateral lung was identified by following the arterial branch emerging from the bifurcation of the main pulmonary artery (Figure 1b); its location was generally apical and medial in the thorax near the fetal heart. The intrapulmonary branches were identified by their characteristic waveform according to Laudy et al The sample volume (1 3 mm) was placed over the most proximal branch of the pulmonary artery. The trim was adjusted to run at 1 cm/s, allowing clear separation of the different components of the waveform including at least three complete cycles. The following pulsed parameters were measured in each branch: pulsatility index (PI), peak systolic velocity (PSV), acceleration time (AT), peak early-diastolic reversed flow (PEDRF), end-diastolic velocity (EDV), time averaged velocity (TAV), systolic time (ST), diastolic time (DT) and AT/ejection time ratio (AT/ET). At least three consecutive good-quality waveforms were measured and the mean was considered as the representative value. values from fetuses with CDH were transformed into Z-scores and analyzed in relation to the. For reproducibility analyses PI values from 40 normal fetuses were collected twice by two operators. Statistical analysis Descriptive statistics were used for all parameters. To construct reference intervals the method described by Royston and Wright 16 was used. Least-squares regression analysis was applied to estimate both the mean and SD curves as polynomial functions of GA. Linear regression was used when the SD of the unscaled residuals appeared to show no trend with GA; otherwise a polynomial regression analysis was used. Reproducibility was evaluated with intra- and interobserver intraclass correlation coefficients (ICCs), and reported with CI. The association between measurements and the O/E LHR was tested with Spearman s rank correlation analysis. Differences between arterial values in the ipsilateral and contralateral lungs were assessed with the Wilcoxon signed rank test and P < 0.0 was considered statistically significant. Differences between different O/E LHR groups were evaluated with the Kruskal Wallis rank test. All analyses were performed with the SPSS 1.0 statistical package (SPSS Inc., Chicago, IL, USA).

3 166 Moreno-Alvarez et al. Figure 1 Localization and waveforms of the main branch of the intrapulmonary artery in contralateral (a) and ipsilateral (b) lungs of fetuses with congenital diaphragmatic hernia. PEDRF, peak early-diastolic reversed flow; PSV, peak systolic velocity. RESULTS Characteristics of the study group are presented in Table 1. All studied cases were born alive. Six of nine neonatal deaths occurred during the first 48 h after birth due to severe pulmonary hypertension. The remaining three infants underwent extracorporeal membrane oxygenation with no clear recovery of respiratory function. One of these cases died 83 days after delivery. Feasibility All parameters were registered in at least one lung of all normally grown fetuses, and in the contralateral lungs from all CDH fetuses. recordings from the ipsilateral lung were collected from 28 CDH fetuses (77.7%). Failure to acquire the waveforms in this lung was mainly due to fetal position, maternal obesity and noise contamination produced by heart movements. Table 1 Characteristics of the study population Number of cases 36 Maternal age (years, median (range)) 32 (23 37) Gestational age at study (weeks + days, 2 + ( to ) median (range)) LHR (mean (range)) 1.11 ( ) Observed/expected LHR (%, mean 33.6 ( ) (range)) Survivors (n) 27 Non-survivors (n) 9 Successful recordings of the 36 (100) contralateral lung (n (%)) Successful recordings of the ipsilateral 28 (77.7) lung (n (%)) LHR, lung-to-head ratio. Reference values for gestational age In the normally grown fetuses the right lung was evaluated in 78 (62%) and the left lung in 47 (38%)

4 in congenital diaphragmatic hernia 167 Table 2 Reference values of distinct parameters in the proximal arterial branch of the intrapulmonary circulation throughout gestation obtained from 12 normal fetuses parameter Regression equation r 2 P PI P0 th : Y = GA 0.00 GA SD: 0.7 PSV (cm/s) P0 th : Y = GA GA SD: 14.0 EDV (cm/s) P0 th : Y = GA GA SD: 2.44 PEDRF (cm/s) P0 th : Y = GA 0.01 GA < SD: Y = GA GA 2 AT (s) P0 th : Y = GA E GA < SD: Y = GA GA 2 TAV (cm/s) P0 th : Y = GA SD: 4.39 AT/ET P0 th : Y = GA GA < SD: AT, acceleration time; AT/ET, acceleration time/ejection time ratio; EDV, end-diastolic velocity; GA, gestational age; P0 th, 0 th percentile; PEDRF, peak early-diastolic reversed flow; PI, pulsatility index; PSV, peak systolic velocity; TAV, time averaged velocity. cases. No statistically significant differences were found in any of the parameters between the two lungs. The resulting formulas for calculating the different measurements in relation to GA are presented in Table 2. PI values showed a steady decrease during pregnancy. All velocities increased with GA but plateaued at about 28 weeks of gestation, whereas AT and the AT/ET ratio increased. Reproducibility analysis showed an intraobserver ICC of 0.87 (9% CI, ) and an interobserver ICC of 0.82 (9% CI, ). parameters in congenital diaphragmatic hernia Differences between the two lungs in the measurements expressed as Z-scores are presented in Table 3. PI, PSV and TAV were significantly increased Table 3 Z-scores of pulsed parameters in the proximal arterial branch of the intrapulmonary circulation of the ipsilateral and contralateral lungs in fetuses with congenital diaphragmatic hernia parameter Contralateral lung (n = 36) Z-score (SD) Ipsilateral lung (n = 28) P PI 1.48 (1.6) 2.2 (1.26) PSV 0.2 (0.99) 1.28 (1.02) EDV 1.14 (1.07) 1.60 (0.82) PEDRF 2.10 (1.98) 0.8 (2.2) AT 0.0 (0.64) 0.3 (0.63) ST 0.19 (1.2) 0.8 (1.99) DT 0.79 (0.93) 0.47 (1.04) TAV 0.42 (0.44) 0.98 (0.44) < AT/ET 0.01 (0.70) 0.37 (0.71) AT, acceleration time; AT/ET, acceleration time/ejection time ratio; DT, diastolic time; EDV, end-diastolic velocity; PEDRF, peak early-diastolic reversed flow; PI, pulsatility index; PSV, peak systolic velocity; ST, systolic time; TAV, time averaged velocity. in the ipsilateral lung, whereas the remaining parameters were similar in both lungs. Plots of the PI and PEDRF values from all studied fetuses are presented in Figure 2. (a) Pulsatility index (b) PEDRF (cm/s) Gestational age (weeks) Gestational age (weeks) Figure 2 (a) Pulsatility index and (b) peak early-diastolic reversed flow (PEDRF) from 20 to 3 weeks of gestation for controls ( ) and from the contralateral ( ) and ipsilateral ( ) lung arteries of fetuses with congenital diaphragmatic hernia. Reference lines ( th, 0 th and 9 th centiles) are shown.

5 168 Moreno-Alvarez et al. (a) 4 Table 4 Spearman s rank correlation coefficients (r) between the observed to expected lung-to-head ratio and pulsed parameters in the proximal arterial branch of the intrapulmonary circulation of the ipsilateral and contralateral lungs in fetuses with congenital diaphragmatic hernia PI Z-score 3 2 Correlation coefficients parameter Contralateral Ipsilateral (b) PI PSV EDV PEDRF AT ST DT TAV AT/ET PEDRF Z-score Figure 3 Mean (SD) Z-scores for (a) pulsatility index (PI) and (b) peak early-diastolic reversed flow (PEDRF) in the four observed to expected lung-to-head ratio (O/E LHR) groups. Group 1, n = 3 (O/E LHR < 1%); Group 2, n = 6 (O/E LHR 16 2%); Group 3, n = 12 (O/E LHR 26 3%); and Group 4, n = 1 (O/E LHR > 36%) 7. Contralateral ( ) and ipsilateral ( ) values are shown. Correlation of parameters with O/E LHR The complete Spearman rank correlation analysis between each parameter and the O/E LHR is described in Table 4. In the contralateral lung all parameters except for ST showed a negative association with O/E LHR, with the highest values observed for PEDRF (r = 0.6) and PI (r = 0.). In the ipsilateral lung EDV, DT and AT/ET showed a positive trend whereas the remaining parameters showed a negative trend with O/E LHR; this relationship was significant only for PEDRF (r = 0.43) (Figure 3). DISCUSSION The results of this study show that the intrapulmonary circulation can be evaluated in fetuses with CDH, that parameters are abnormal in almost all fetuses with CDH, and that there is a statistically significant correlation between LHR and mainly PEDRF, suggesting a potential use for evaluation in a composite prognostic score. AT, acceleration time; AT/ET, acceleration time/ejection time ratio; DT, diastolic time; EDV, end-diastolic velocity; PEDRF, peak early-diastolic reversed flow; PI, pulsatility index; PSV, peak systolic velocity; ST, systolic time; TAV, time averaged velocity. Although changes in the intrapulmonary circulation have been reported previously in association with a heterogeneous group of conditions carrying a risk of pulmonary hypoplasia 9,11, we focused only on fetuses with CDH at different stages of lung underdevelopment. A high success rate was achieved in obtaining signals even from the lung ipsilateral to the CDH defect. The different values observed between the two lungs may be associated with changes in vessel diameter 17 and in peripheral resistance. Possible reasons for unsuccessful recording of the ipsilateral lung include the extended deviation of this lung, its proximity to the main pulmonary artery, and its closeness to the heart. The strongest correlation with O/E LHR was found with PEDRF in both lungs, and with PI in the contralateral lung only. PEDRF is characteristic of the intrapulmonary artery waveform and can be explained by the high resistance of the vascular pulmonary tree and by preferential blood flow through the ductus arteriosus 13. Normal reference values of PEDRF throughout gestation have been reported previously by Laudy et al. 14. In concordance with their results, we found that PEDRF increased with GA. The increment in PEDRF in fetuses with a low O/E LHR might be explained by the increase in intravascular pressure. The clinical significance of PEDRF has been explored previously by Chaoui et al. 18, who reported no changes in this parameter in fetuses which subsequently died neonatally with histological corroboration of lethal pulmonary hypoplasia. Our results show that the strong association between PEDRF and O/E LHR could be a complementary parameter to investigate in cases where there is a discrepancy in the LHR estimated by different operators. PI values showed a reduction in relation to the O/E LHR, but almost all values from Groups 3 and 4 were within the normal range, limiting the ability of the PI

6 in congenital diaphragmatic hernia 169 to identify the severity of lung area restriction. Increased pulsatility might be a direct consequence of the reduced vascular tree in relation to the stroke volume of a normal right ventricle. Pulmonary blood flow represents approximately 1 20% of the cardiac output 19 and any increase in resistance in the pulmonary circuit will result in a shift of flow towards the ductus arteriosus. For these reasons, PI is probably only an indirect estimate of hypoplastic changes. Evaluation of PEDRF may offer additional information about peripheral resistance and of the flow that is shifted to the descending aorta through the ductus arteriosus. The strong association between LHR and PEDRF could improve the prognostic value of the LHR, thus reinforcing its clinical value. In O/E LHR Group 1 (poorest prognosis), PEDRF and PI were significantly abnormal. However, in cases with a moderate to good prognosis the only parameter with discriminatory power was PEDRF. PEDRF might be an indicator of vascular development in the lung and an indirect estimator of bronchial branching, as well as a complementary parameter for the clinical application of the LHR. This study is limited by the fact that we did not directly correlate measurements with survival. The decision of the parents to terminate the pregnancy in some early recognized CDH cases prevented us from including more cases. However, the main purpose of this study was to evaluate the feasibility of imaging and to explore its potential use with the LHR. In large clinical series the LHR has been demonstrated to predict mortality mainly in fetuses with liver herniation 6. In CDH fetuses the comparison between observed LHR and that expected for GA aims to improve the predictive capacity for pulmonary hypoplasia. In CDH fetuses with an O/E LHR between 26 and 3% the survival rate is 6%, meaning that in the best conditions almost 4/10 cases will have a poor perinatal outcome. The size of the contralateral lung area measured to determine the LHR is not solely dependent on the real size of the lung, but also depends on the variable deformation of the lung by viscerated abdominal organs 20. As a result a differing degree of overestimation of the real size of the lung is inherent in the measurement of the LHR. To address this, other alternatives such as investigations and three-dimensional ultrasound lung volume estimation 21 might improve the prediction of pulmonary hypoplasia. In summary, the value of assessment in CDH may be twofold. First, it could constitute an additional parameter in composite scoring for evaluation of prognosis, reducing the individual variability of LHR. Second, it may provide relevant information on different aspects relating to prenatal therapy of CDH 22. Tracheal occlusion has a relatively predictable response on the basis of LHR, but a small proportion of cases fail to respond to therapy for unknown reasons. Therefore, LHR alone cannot fully predict pulmonary response after tracheal occlusion, and indices might represent additional predictive parameters. Likewise, these parameters could be useful in documenting lung response after intrauterine therapy, thus providing information of critical importance in clinical decision-making. The potential contribution of measurements for these purposes is now under evaluation. ACKNOWLEDGMENTS O.M.-A. was supported by research grants from the Instituto Mexicano del Seguro Social (IMSS) and Consejo Nacional de Ciencia y Tecnología (CONACyT), Mexico. He also thanks the UMAE Luis Castelazo Ayala for supporting his predoctoral stay at the Hospital Clinic in Barcelona, Spain. E.H.-A. was supported by research grants from the Ministry of Education and Science (SB ), Juan de la Cierva Program for the Development of Research Centers, Spain. He expresses his gratitude to the National Institute of Perinatal Medicine (INPer) in Mexico. REFERENCES 1. Stege G, Fenton A, Jaffray B. Nihilism in the 1990s. The true mortality of CDH. Pediatrics 2003; 112: Adzick NS, Vacanti JP, Lillehei CW, O Rourke PP, Crone RK, Wilson JM. Fetal diaphragmatic hernia: ultrasound diagnosis and clinical outcome in 38 cases. J Pediatr Surg 1989; 24: Harrison MR, Adzick NS, Estes JM, Howell LJ. A prospective study of the outcome for fetuses with diaphragmatic hernia. JAMA 1994; 271: Greenough A. Factors adversely affecting lung growth. Paediatr Respir Rev 2000; 1: Laudy JA, Wladimiroff JW. The fetal lung 1: developmental aspects. Ultrasound Obstet Gynecol 2000; 16: Jani J, Keller R, Benachi A, Nicolaides KH, Favre R, Gratacos E, Laudy J, Eisenberg V, Eggink A, Vaast P, Deprest J; Antenatal-CDH-Registry Group. Prenatal prediction of survival in isolated left-sided diaphragmatic hernia. Ultrasound Obstet Gynecol 2006; 27: Jani J, Nicolaides KH, Keller RL, Benachi A, Peralta CFA, Favre R, Moreno O, Tibboel D, Lipitz S, Eggink A, Vaast P, Allegaert K, Harrison M, Deprest J; Antenatal-CDH-Registry Group. Observed to expected lung area to head circumference ratio in the prediction of survival in fetuses with isolated diaphragmatic hernia. Ultrasound Obstet Gynecol 2007; 30: Kotecha S. Lung growth for beginners. Paediatr Respir Rev 2000; 1: Laudy JA, Tibboel D, Robben SG, de Krijger RR, de Ridder MA, Wladimiroff JW. Prenatal prediction of pulmonary hypoplasia: clinical, biometric, and velocity correlates. Pediatrics 2002; 109: Fuke S, Kanzaki T, Mu J, Wasada K, Takemura M, Mitsuda N, Murata Y. Antenatal prediction of pulmonary hypoplasia by acceleration time/ejection time ratio of fetal pulmonary arteries by blood flow velocimetry. Am J Obstet Gynecol 2003; 188: Yoshimura S, Masuzaki H, Miura K, Muta K, Gotoh H, Ishimaru T. Diagnosis of fetal pulmonary hypoplasia by measurement of blood flow velocity waveforms of pulmonary arteries with ultrasonography. Am J Obstet Gynecol 1999; 180: Metkus AP, Filly RA, Stringer MD, Harrison MR, Adzick NS. Sonographic predictors of survival in fetal diaphragmatic hernia. JPediatrSurg1996; 31:

7 170 Moreno-Alvarez et al. 13. Laudy JA. ultrasonography of the human fetal pulmonary circulation. Eur J Obstet Gynecol Reprod Biol 2001; 99: Laudy JA, de Ridder MA, Wladimiroff JW. velocimetry in branch pulmonary arteries of normal human fetuses during the second half of gestation. Pediatr Res 1997; 41: Laudy JAM, de Ridder MA, Wladimiroff JW. Human fetal pulmonary artery velocimetry: repeatability and normal values with emphasis on middle and distal pulmonary vessels. Ultrasound Obstet Gynecol 2000; 1: Royston P, Wright EM. How to construct normal ranges for fetal variables. Ultrasound Obstet Gynecol 1998; 11: Sokol J, Bohn D, Lacro RV, Ryan G, Stephens D, Rabinovitch M, Smallhorn J, Hornberger LK. Fetal pulmonary artery diameters and their association with lung hypoplasia and postnatal outcome in congenital diaphragmatic hernia. Am J Obstet Gynecol 2002; 186: Chaoui R, Kalache K, Tennstedt C, Lenz F, Vogel M. Pulmonary arterial velocimetry in fetuses with lung hypoplasia. Eur J Obstet Gynecol Reprod Biol 1999; 84: Rudolph AM. Distribution and regulation of blood flow in the fetal and neonatal lamb. Circ Res 198; 7: Jani J, Peralta CF, Van Schoubroeck D, Deprest J, Nicolaides KH. Relation between lung-to-head ratio and lung volume in normal fetuses and fetuses with diaphragmatic hernia. Ultrasound Obstet Gynecol 2006; 27: Ruano R, Benachi A, Joubin L, Aubry MC, Thalabard JC, Dumez Y, Dommergues M. Three-dimensional ultrasonographic assessment of fetal lung volume as prognostic factor in isolated congenital diaphragmatic hernia. BJOG 2004; 111: Deprest J, Gratacos E, Nicolaides KH; FETO Task Group. Fetoscopic tracheal occlusion (FETO) for severe congenital diaphragmatic hernia: evolution of a technique and preliminary results. Ultrasound Obstet Gynecol 2004; 24: