Objective
Congenital diaphragmatic hernia (CDH) is clinically challenging because of associated lung hypoplasia (LH). There have been no validated parameters to evaluate fetal LH severity. Sildenafil has been shown to improve LH mass in nitrofen-induced pulmonary artery (PA) models, but the pulmonary vascular tone has not been evaluated in vivo. The aim of this study was to identify the PA Doppler parameter that best predicts LH severity and to investigate the efficacy of antenatal sildenafil treatment in experimental CDH.
Study Design
Nitrofen (50-60% CDH in offspring) or vehicle on E9.5 and sildenafil or vehicle on E11.5-20.5 were administrated to pregnant rats. On E20.5, PA Doppler indices were investigated with and without maternal hyperoxia. The presence/absence of CDH, lung/body weight ratio and radial saccular count were assessed at E20.5.
Results
At baseline, CDH rats had lower PA Doppler acceleration/ejection time ratios and pulsatility index (PI). Maternal hyperoxia resulted in a significant decrease in the PA/PI suggesting pulmonary vasodilation. In contrast, in CDH fetuses, the ipsilateral PA/PI showed little or no response to hyperoxia ( P > .05), and in those with LH, PI response to maternal hyperoxia correlated positively with hernia, lung/body weight ratio (r = 0.70, P = .01). Maternal sildenafil therapy significantly improved PA response to hyperoxia and lung growth in CDH fetuses ( P < .01).
Conclusion
Pulmonary vasodilation that occurs in E20.5 fetal rats in response to maternal hyperoxia is blunted in CDH. Change in PA/PI with hyperoxia is a useful predictor of LH severity. Sildenafil improves pulmonary vascular response and lung growth in fetal CDH.
Congenital diaphragmatic hernia (CDH) is a clinically challenging malformation with high neonatal mortality because of associated lung hypoplasia (LH) despite significant advances in perinatal and neonatal care. Several investigators have described ultrasound indices in fetal CDH predictive of outcome that could potentially improve prenatal counseling and planning of perinatal management, as well as identify the most appropriate group for fetal intervention. Initial ultrasound-based parameters of LH severity described included degree of cardiac shift, presence of liver herniation, and lung area to head circumference ratio (LHR). Although used routinely in many practices, not one of these variables has been consistently shown to predict clinical CDH outcome. More recently, branch pulmonary artery (PA) diameters have been shown to correlate with not only postmortem lung weight but also respiratory morbidity in fetal CDH. Doppler-based PA flow parameters have also been used to assess LH indirectly by their reflection of pulmonary vascular resistance. Shorter acceleration time (AT) to ejection time (ET) ratio, previously documented in adults with pulmonary hypertension, has been observed in fetuses with neonatal mortality because of LH from various causes. Greater peak early-diastolic reverse flow (PEDRF) observed in fetuses with CDH and diminutive LHR, has been suggested to reflect high pulmonary vascular resistance with preferential ductus arteriosus flow in diastole. Finally, lack of response to maternal hyperoxia using the PA pulsatility index (PI) (=(Peak systolic velocity — end diastolic velocity)/mean velocity) after 32 weeks’ gestation has also been identified as a possible predictor of LH severity. Neither fetal branch PA diameters nor flow parameters with and without maternal hyperoxia have been validated as accurate predictors of postnatal LH severity. Furthermore, innumerable confounding factors including additional pathology, infection, and even management styles may contribute to difficulties in evaluating indices of LH severity in the clinical setting.
Nitrofen administrated to maternal rat at E9.5 induces a spectrum of LH in most and CDH in up to 50-60% of offspring, and is a well-established model for the investigation of CDH. In this model, CDH is associated with both reduced lung mass and PA vasoreactivity ex vivo. Maternal treatment beginning at E11.5 with sildenafil, a potent and highly selective inhibitor of phosphodiesterase (PDE) 5, has been shown recently to result in improved lung growth and improved pulmonary vascular reactivity ex vivo in CDH fetuses. Further in vivo data to support the efficacy of antenatal sildenafil treatment is required before it should be considered for use in clinical trials.
The objectives of the present study were 2-fold: (1) to determine the most accurate fetal echocardiographic parameter for prediction of LH severity using the nitrofen-induced CDH rat model, and (2) to evaluate the influence of prenatal sildenafil treatment using these parameters in this model.
Materials and Methods
Animal model
This study was approved by the animal care and use committee at the University of Alberta. Pregnant Sprague-Dawley rats were fed 100 mg of nitrofen dissolved in 1 mL olive oil or vehicle on E9.5. Sildenafil (100 mg/kg/d) or vehicle was administered (subcutaneous injection) from E11.5 to E20.5 as previously described. As nitrofen induces CDH in 50-60% of the offspring of exposed maternal rats, pregnant rats were randomly assigned to 6 fetal groups at the stage of ordering in a 2-way factorial design: (1) control, (2) nitrofen-induced CDH, (3) nitrofen treated without CDH, (4) sildenafil control (5) nitrofen-induced CDH + sildenafil, (6) nitrofen treated without CDH + sildenafil. The individual performing the echocardiographic assessments and offline measures was blinded to the treatment group. Treatment groups were only unmasked after all data was acquired.
Fetal echocardiography
On E20.5 (at term), 8-10 hours from the last dose of sildenafil, fetal echocardiography was performed under anesthesia with 1.0-1.5% isoflurane, known not to impact cardiac function in the rat using the Vevo 770 or Vevo 2100 ultrasound biomicroscope (Visualsonics, Toronto, Canada). After defining all fetal positions, 3 fetuses on either side of the maternal bladder were evaluated. An image of the 3 vessel view equivalent that demonstrated the main and branch pulmonary arteries, ascending aorta and superior vena cava in cross-section was first acquired ( Figure 1 ). Branch PA diameters were measured just after their takeoff from the main PA from a plane that was axial to the lumen. Doppler interrogation of both branch PAs was attempted with an angle of incidence of <20° to be as parallel to the direction of blood flow as possible. Doppler spectral tracing including AT/ET, PEDRF, and PI were evaluated without and with maternal hyperoxia ( Figure 2 , A and B). For the maternal hyperoxia challenge, the pregnant rats were exposed to 100% FiO 2 by nasal catheter for 10 minutes followed by repeated Doppler assessment of the same fetal rat. After each fetus was assessed, the maternal supplemental O 2 was discontinued for 10 minutes, permitting a return to baseline before the evaluation of the next fetus. The Doppler assessments were conducted 3 times for each evaluation and averaged for the analysis. The effect of maternal hyperoxia was calculated as a percentage; (the value with hyperoxia-without hyperoxia)/value without hyperoxia.
Lung growth assessment
On completion of fetal echo assessments, the fetuses were harvested by caesarian section so as to identify the fetuses in which the in vivo data was acquired. The presence or absence of CDH, lung weight (bilateral lungs)/body weight ratio (LW/BW) and radial saccular count (RSC) were assessed. LW/BW is an established marker of severity of LH. The lungs were fixed in 4% paraformaldehyde overnight and embedded in paraffin, cut into 3 μm sections, and stained with hematoxylin and eosin (H&E). The stained sections were analyzed with OpenLab Imaging System (Quorum Technologies Inc., Guelph Ontario, Canada). The RSC was estimated as previously described.
Statistical analysis
Statistical assessment was performed with GraphPad Prism software (version 5.0a; Graphpad, La Jolla, CA). The values were expressed as mean ± standard deviation (SD) and statistical significance was defined as P < .05. Comparison between 2 groups was performed using unpaired Student t test. One-way ANOVA was used to compare multiple groups with posthoc testing with Tukey’s multiple comparison test. The correlations between LW/BW and lung parameters were evaluated with using Pearson’s correlation test.
Results
Thirty maternal rats were divided into 6 fetal groups as shown above. Based on 2 dimensional analysis, the branch PA diameters were smaller in nitrofen-induced CDH (+) compared with control and nitrofen-CDH (-) ( Table 1 ; P < .05) The RPA was easier to more consistently measure in the fetal rats, perhaps because of its visibility in the 3 vessel view, in contrast to the LPA which courses just beneath the ductus arteriosus ( Figure 1 ). Nitrofen-induced CDH (+) fetal rats had significantly smaller LW/BW compared with control and nitrofen-CDH (−) ( P < .05) ( Table 1 ). As LW/BW was measured including both lungs and not separate left and right lungs, we were unable to correlate branch PA diameter with this variable to determine its accuracy in predicting LH severity.
| Variable | Control | Nitrofen (+) CDH (+) | Nitrofen(+) CDH (−) | P value a |
|---|---|---|---|---|
| n | 36 | 20 | 41 | |
| LW/BW | 0.031 ± 0.003 | 0.026 ± 0.007 a | 0.029 ± 0.007 | < .05 |
| AT/ET ratio | 0.093 ± 0.010 | 0.084 ± 0.015 a | 0.087 ± 0.097 | < .05 |
| PEDRF | 18.64 ± 6.000 | 15.00 ± 6.334 | 18.63 ± 6.326 | .088 |
| PI | 2.090 ± 0.122 | 1.924 ± 0.147 a | 2.021 ± 0.177 | < .001 |
| PA, mm | 0.36 ± 0.05 | 0.33 ± 0.05 a | 0.34 ± 0.04 | < .05 |
a Statistically significant differences were only observed between controls and nitrofen+/CDH+ pups, not between controls and nitrofen+/CDH− pups.
In vivo Doppler interrogation of late fetal branch PA flow
The branch PA Doppler spectral tracing obtained in the E20.5 fetal rats is demonstrated in Figure 1 . First, the 3 vessel view could be clearly identified ( Figure 1 ) and the pulsed Doppler signal was also comparable to that observed in human fetuses ( Figure 2 , C). Doppler interrogation of the branch PAs was facilitated in fetuses with CDH by identifying flow in the pulmonary veins ( Figure 2 , D) (spectra also comparable to those obtained in humans) and sweeping cephalad during the interrogation until the characteristic PA flow signal was observed.
At baseline, AT/ET ratio, PEDRF and PI did not differ between control and nitrofen-exposed CDH (−) fetal rats, whereas nitrofen-induced CDH (+) rats had significantly lower AT/ET ( P < .05) and PI ( P < .001) ( Table 1 ). Secondly, in response to maternal hyperoxia, we observed no significant change in AT/ET and PEDRF, but PI significantly decreased in control rats ( P < .01). In nitrofen-exposed fetal rats, nitrofen-CDH (−) rats showed a comparable decrease in PA PI to controls ( P < .001) whereas nitrofen-CDH (+) rats had on average no significant response in PI to maternal hyperoxia ( Figure 3 ). In nitrofen-CDH (+) fetal rats, although the PA contralateral to the CDH did respond similarly to that of controls, the ipsilateral PA showed no significant response to maternal hyperoxia ( Figure 4 , A; P < .05). Finally, the responsivity of PI to maternal hyperoxia in fetal rats with hypoplastic lungs, defined as LW/BW of less than 2 standard deviations below the mean observed in normals, positively correlated with LW/BW in fetal rats with hypoplastic lungs ( P = .01, r = −0.70) ( Figure 4 , B). No other fetal echo parameter correlated with LW/BW among pups with lung hypoplasia defined as LW/BW of less than 2 standard deviations below the mean observed in normals.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
