| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Clinical Radiology, University Hospital Mannheim, University of
Heidelberg, Theodor Kutzer Ufer 1-3, 68167 Mannheim, Germany.
2 Department of Pediatrics, University Hospital Mannheim, University of
Heidelberg, Mannheim, Germany.
Received April 17, 2007;
accepted after revision June 10, 2007.
Address correspondence to K. W. Neff
(wolfgang.neff{at}rad.ma.uni-heidelberg.de).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. The fetal lung volume measurements of 65 fetuses with CDH were obtained between 32 and 34 weeks' gestation by means of MRI performed with multiplanar T2-weighted HASTE and true fast imaging with steady-state precession sequences. Logistic regression analysis was used to assess the prognostic value of the fetal lung volume measurements for prenatal prediction of fetal survival and need for neonatal ECMO.
RESULTS. Fetal lung volume was a highly significant predictor of survival (p < 0.0001) and neonatal ECMO requirement (p = 0.0006). The mortality was 84% and the ECMO requirement 80% among fetuses with a lung volume of 5 mL. The mortality was 0.4% and the ECMO requirement 20% among patients with a fetal lung volume of 30 mL.
CONCLUSION. Logistic regression analysis of MRI fetal lung volume measurements is highly valuable in predicting mortality among neonates with CDH, and it may help to estimate the need for neonatal ECMO. The method is feasible for facilitating parental guidance and may help in choosing postnatal therapeutic options, including ECMO therapy.
Keywords: congenital diaphragmatic hernia • ECMO • extracorporeal membrane oxygenation • fetal lung volume measurement • MRI
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The overall prenatal mortality of CDH is influenced by the rate of elective pregnancy termination [1, 2]. Among the population assessed by Colvin et al. [1], pregnancy was terminated in 49% of cases of prenatal diagnosis. In cases of postnatal diagnosis, however, the degree of pulmonary hypoplasia and associated abnormalities has the greatest influence on survival [5–7]. These conditions usually lead to death within the first 24 hours of life [2]. Although several surgical referral centers have reported survival rates up to 93% [1, 8–10], the results may have been influenced by patient selection bias. Colvin et al. reported that 35% of live-born infants died before referral and that the population of infants reaching the tertiary surgical center represented only 40% of the total cases of CDH.
A specialist involved in the care of a patient with CDH may find it difficult to provide detailed prognostic information and to individually counsel parents on therapeutic options. Few parameters have been defined to aid in predicting outcome for these infants. Although a variety of indexes have been suggested, no robust marker is available for accurate prediction of patient survival. Attention has been focused on the lung-to-head ratio assessed with sonography [11–15] and relative fetal lung volume calculated with biometric parameters based on MRI findings [16–19]. However, both parameters allow only a rough estimate of likelihood of survival. To our knowledge, no method is available for individually determining survival of infants with CDH. There also is no parameter for estimating the need for neonatal extracorporeal membrane oxygenation (ECMO) therapy. Among patients whose condition is refractory to conventional ventilation therapy, ECMO has been reported to improve the survival rate [20–23]. However, ECMO therapy is available at only a few specialized care centers, so prenatal transfer of the patient is desirable. Furthermore, and particularly for critically ill neonates, a generally higher survival rate has been observed at experienced tertiary care centers (high-volume centers) than at institutions where patients with CDH are treated less frequently [24]. Thus, even if ECMO therapy is not advocated, prenatal identification of patients at high risk of neonatal respiratory failure is essential to avoid delay in intensified postnatal care and to improve outcome for these patients.
|
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
ECMO Therapy
All infants were intubated immediately after birth, and gentle conventional
ventilation was administered. ECMO therapy was initiated if the postductal
PaO2 was less than 40 mm Hg and preductal saturation did not rise
above 80 mm Hg for more than 2 hours or if the postductal PaO2 did
not rise above 50 mm Hg and preductal saturation stayed below 95% for more
than 4 hours. During ECMO, heparin (Liquemin N5000, Roche) was given IV at a
daily dose of 400 IU/kg. At decannulation the jugular vein was ligated, and
the common carotid artery was reconstructed. Exclusion criteria for
institution of ECMO therapy were evidence of ongoing bleeding or severe
coagulopathy, fatal concomitant anomalies, and lactate levels of 20 mmol/L or
greater.
MRI Planimetry
All MR images were obtained with a 1.5-T supraconducting MRI system
(Magnetom Sonata or Avanto, Siemens Medical Solutions) with a six-element
phased-array surface coil. The mothers were positioned in either the supine or
the partial left lateral position. No sedation was given to reduce fetal
movement. The imaging protocol consisted of multiplanar T2-weighted images
without respiratory triggering. A HASTE sequence (TR/TE, 1,000/85; flip angle,
150°; matrix size, 512 x 512) was performed on all patients. From
June 2005 onward we also conducted a multiplanar true fast imaging with
steady-state precession (true FISP) sequence on all of the patients (4.3/1.9;
flip angle, 59°; matrix size, 512 x 512). Because its acquisition
time (
14 seconds) was shorter than that of the HASTE sequence ( 22
seconds), the true FISP sequence was particularly valuable if the fetus was
agitated. All images were obtained with a 4-mm slice thickness. Sections were
adjusted to the transverse, coronal, and sagittal planes relative to the fetal
lungs. Imaging was monitored by an experienced radiologist to ensure that all
of the anatomic features of interest were included. Sequences degraded by
fetal motion artifacts were repeated to obtain images covering the whole
thorax in a single acquisition and allowing clear identification of parietal
and mediastinal boundaries.
Lung volumes were measured by two investigators using volume analysis software (ARGUS, Siemens Medical Solutions) on a workstation (Leonardo, Siemens Medical Solutions). A hand-tracing drawing tool was used to outline the region of interest following the lung boundaries on consecutive images covering the entire thorax (Fig. 1A, 1B, 1C, 1D). A computer algorithm was used to calculate the area of the region of interest in square millimeters. The area was multiplied by the slice thickness to obtain the volume of the entire lung. When different section orientations of high image quality were available, the best images were analyzed, and the mean fetal lung volume was calculated for subsequent evaluation.
Data Analysis
To assess the effect of fetal lung volume, maternal age, and the side of
the defect on survival and neonatal ECMO requirement, we applied the Fisher's
exact test and the Student's t test. Logistic regression analysis was
used to express the association between mean fetal lung volume calculated with
MR planimetry and mortality and likelihood of need for neonatal ECMO therapy.
Statistical calculations were performed with SAS release 8.02 (SAS Institute).
A value of p < 0.05 was considered to indicate a significant
difference.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Patient Survival and Neonatal ECMO Requirement
The influence of certain factors on survival and the need for ECMO are
presented in Table 1. The
global survival rate was 83%. The location of the defect in neonates who
survived did not differ from that in infants who did not. Among the 11
patients who died, including the patient with a bilateral defect, eight deaths
occurred because of respiratory failure, and three infants died of major
hemorrhage during ECMO therapy (two cases of intracerebral and one of cervical
bleeding, 11% of all ECMO patients). The mean fetal lung volume of the infants
whose deaths were related to ECMO did not differ significantly from that of
infants who died of respiratory failure (10.5 ± 6.4 mL vs 12.1 ±
5.2 mL, p = 0.610). The fetal lung volume calculated with prenatal
MRI was the only parameter that significantly influenced infant survival
(p < 0.0001). The average lung volume of survivors was 23.5
± 8.8 mL, whereas that of nonsurvivors was 11.7 ± 5.3 mL.
Because of respiratory distress, 27 (42%) of the newborns received arteriovenous ECMO therapy. The fetal lung volume measured on prenatal MRI again was found to be the only predictive parameter for estimating postnatal course. Higher fetal lung volume was significantly associated with a lower neonatal ECMO requirement (p = 0.005). All other parameters, including location of the hernia, were of no consequence.
The association between individual fetal lung volume and postnatal
mortality (Rmortality) and neonatal ECMO requirement
(RECMO) was determined using logistic regression analysis
performed with SAS release 2.0 and was expressed as follows:
 |
 |
Application of these formulas indicated an 84% mortality among infants with a prenatal fetal lung volume of 5 mL and a 0.4% mortality among infants with a 30-mL lung volume. The corresponding need for ECMO therapy was estimated at 80% and 20%, respectively. Table 2 and Figure 2 provide an overview of the data.
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The lung-to-head ratio measured with sonography has been proposed as a reliable means of helping to predict postnatal survival. In isolated left-sided CDH, the lung-to-head ratio is the product of the orthogonal diameters of the right lung at the level of the cardiac atria divided by the head circumference. Several prospective and retrospective studies have shown that cutoff levels of lung-to-head ratio of 1 and 1.4 can be applied in predicting fetal outcome. The predicted survival rate is 100% for a lung-to-head ratio greater than 1.4, and the predicted mortality is 100% for a lung-to-head ratio less than 1 [11–13, 33]. A considerable number of these patients, however, have intermediate values, and our ability to predict outcome among these patients is limited, survival rates ranging from 38% to 61% [11, 13, 33]. Jani et al. [34] described a detailed association between lung-to-head ratio and postnatal outcome, the survival rate ranging from 17% to 78%. The data in that study, however, were limited to fetuses in early pregnancy (25–29 weeks' gestation) with associated liver herniation and a lung-to-head ratio of only 0.9 or less. Other studies [14, 35, 36] did not confirm a predictive value of the lung-to-head ratio for either survival or postnatal clinical course, including the need for ECMO therapy.
Relative fetal lung volume assessed with prenatal MR planimetry has been described as a prognostic marker. Gorincour et al. [17] observed marked impairment in the survival rate among infants with a relative fetal lung volume less than 25% of the expected volume. Those authors, however, did not provide more accurate calculation of individual survival likelihood, nor did investigators in subsequent studies on this topic [16–19].
The first aims of our study were to evaluate a new technique of prenatal estimation of individual likelihood of survival among infants with isolated CDH and to determine the corresponding probability of need for ECMO therapy. Using logistic regression analysis, we found that fetal lung volume assessed with prenatal MR planimetry at 32–34 weeks' gestation is significantly associated with survival and need for neonatal ECMO. These findings show that fetal lung volume assessed with MR planimetry can be used for reliable estimation of the postnatal course of each patient. MRI fetal lung volume measurement may even be preferable to lung-to-head ratio as a predictive parameter because measurement accuracy and reliability have been shown to be high, with a low intraobserver and interobserver variability [37]. Moreover, imaging quality is almost unaffected by maternal obesity, oligohydramnios, and an unfavorable fetal position.
The main limitation of our study was that we restricted the patient cohort to fetuses between 32 and 34 weeks' gestation, that is, late pregnancy. However, we imposed this restriction deliberately to first assess the prognostic power of the method free from a potential effect of gestational age. Our study was designed not to identify infants who might benefit from prenatal high-risk interventions such as fetal endotracheal balloon implantation [38, 39] but to improve prenatal parental counseling and decisions on postnatal therapy, including ECMO therapy.
Ability to identify fetuses most severely affected in the perinatal period would be invaluable in selecting patients for high-risk interventions. Although an initial randomized trial by Harrison et al. [40] was terminated early by the steering committee because of unexpectedly good outcome in the control group, fetal tracheal occlusion continues to be explored in Europe because further technical refinement may produce survival benefits among patients at high risk [39, 41]. Thus, further evaluation of logistic regression analysis based on MRI fetal lung volume measurements is necessary to assess the prognostic value of this method in early pregnancy to possibly identify fetuses who may benefit from minimally invasive fetal surgery.
Another limitation of our study was that the estimated risk of need for ECMO therapy calculated cannot necessarily be assigned to other centers because there are no universally accepted ECMO criteria. The inclusion and exclusion criteria used at our institution correspond to classic objective criteria for the need to institute ECMO therapy [20, 42, 43]. Nevertheless, procedures may vary among tertiary care centers.
Logistic regression analysis based on MRI fetal lung volume measurements in late pregnancy may serve as a reliable tool to help in prediction of fetal outcome and, consequently, in counseling of clinicians and parents facing difficult decisions in pre- and postnatal care.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. K. Kilian, T. Schaible, V. Hofmann, J. Brade, K. W. Neff, and K. A. Busing Congenital Diaphragmatic Hernia: Predictive Value of MRI Relative Lung-to-Head Ratio Compared with MRI Fetal Lung Volume and Sonographic Lung-to-Head Ratio Am. J. Roentgenol., January 1, 2009; 192(1): 153 - 158. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
