Published online before print April 21, 2005, 10.1148/radiol.2353040280
(Radiology 2005;235:1005-1010.)
© RSNA, 2005
Fetal Lung-to-Liver Signal Intensity Ratio at MR Imaging: Development of a Normal Scale and Possible Role in Predicting Pulmonary Hypoplasia in Utero1
Lee J. Brewerton, BSc,
Radha S. Chari, MD, FRCSC,
Yuanyuan Liang, MSc and
Ravi Bhargava, MD, FRCPC
1 From the University of Alberta Medical School, Edmonton, Alberta, Canada (L.J.B.), and the Departments of Obstetrics and Gynaecology (R.S.C.), Statistical and Mathematical Sciences (Y.L.), Radiology and Diagnostic Imaging (R.B.), and Pediatrics (R.B.), University of Alberta, 2A2.42 Walter C. Mackenzie Health Sciences Centre, 8440 112 St, Edmonton, AB, Canada T6G 2B7. Received February 13, 2004; revision requested April 20; revision received June 10; accepted July 21. Supported by a grant from the Royal Alexandra Hospital Foundation Research Fund and the Dr M. E. Ledingham Memorial Research Award. Address correspondence to R.B. (e-mail: rbhargav@cha.ab.ca).
 |
ABSTRACT
|
|---|
PURPOSE: To define retrospectively a normal range for lung-to-liver signal intensity ratio (LLSIR) in fetuses of 16–40 weeks gestation by using half-Fourier single-shot turbo spin-echo magnetic resonance (MR) imaging.
MATERIALS AND METHODS: Approval from the regional ethics review board for retrospective evaluation was obtained, and informed consent was waived. Retrospective analysis and follow-up of 157 pregnant women who underwent MR imaging over the past 4 years were performed. Seventy-four fetuses were subsequently identified as having clinically normal lung function or normal lung morphologic features at autopsy. A total of 141 normal lungs were analyzed, and the LLSIR was calculated from images on an MR workstation. A mixed-effects statistical model was applied, and 95% prediction intervals were calculated. Ten fetuses with hypoplastic lungs at autopsy were also evaluated.
RESULTS: Plotting LLSIR against gestational age demonstrated that, according to the fitted mean curve, the signal intensity ratio was higher with more advanced gestational age. Statistical modeling suggests a quadratic relationship between gestational age and LLSIR. For fetuses in the normal population, the LLSIR ranged from 1.52 at 21 weeks gestation to 4.31 at 34 weeks gestation. For all hypoplastic lungs in fetuses at or beyond 25 weeks gestation, the LLSIR was outside the lower bound of the 95% prediction interval for the normal population. The distinction between hypoplastic lungs and normal lungs at less than 25 weeks gestation is less definitive.
CONCLUSION: This study provides a normal scale with a 95% prediction interval for LLSIR.
© RSNA, 2005
|
INTRODUCTION
|
|---|
Sufficient pulmonary development in utero is an important determinant of neonatal viability. A number of conditions, such as prolonged oligohydramnios and intrathoracic mass, put a fetus at risk for pulmonary hypoplasia. Lung hypoplasia results in substantial morbidity and mortality after birth. Unfortunately, an effective method of accurately diagnosing pulmonary hypoplasia prenatally has yet to be determined (1). The usefulness of magnetic resonance (MR) imaging in the diagnosis of this condition has been evaluated by a number of groups. MR volume calculations for fetuses with normal lungs have been compared with volume calculations for fetuses with pulmonary hypoplasia and associated conditions, such as congenital diaphragmatic hernia (2–7). In a previous study, investigators concluded that fetal lung volume could be predicted with moderate accuracy by using MR imaging (8).
While the use of MR imaging–derived lung volumes may be useful in identifying some cases of fetal pulmonary hypoplasia, the neonatal and clinical importance of lung volume is still unclear. In one study, the majority of fetuses with oligohydramnios had lung volume ranges that were within the normal lung volume range (5). Paek et al (6) determined that relative lung volume was predictive of outcome in cases of left-sided congenital diaphragmatic hernia, while other investigators determined that lung volume differences in surviving and nonsurviving fetuses were statistically insignificant in cases of congenital diaphragmatic hernia (3). In one study, however, researchers noted two cases of congenital diaphragmatic hernia in which the fetuses had similar lung volumes; in these cases, the surviving fetus had a higher lung-to-liver signal intensity ratio (LLSIR) than the nonsurviving fetus (5).
Signal intensity of the lung at MR imaging has been evaluated in several studies and may be useful in diagnosing pulmonary hypoplasia that could be of neonatal importance (9–11). However, Levine et al (9) performed a qualitative examination of signal intensity and suggested that the use of signal intensity in the diagnosis of pulmonary hypoplasia may not be consistent in the second trimester. Kuwashima et al (10) calculated an LLSIR in 13 fetuses with normal lung function and in 10 fetuses with hypoplastic lungs. They suggested that LLSIR may allow prediction of pulmonary hypoplasia. The signal intensity of a tissue is dependent on its distance from the receiver coil. We believe that a ratio comparing the signal intensity of the lungs with that of another structure of similar depth could provide a correction factor. The liver is adjacent to the lung, has relatively homogeneous signal intensity, and, thus, may serve as a control for depth.
To our knowledge, the validity of MR imaging signal intensity and, specifically, LLSIR in the diagnosis of pulmonary hypoplasia has not been determined because of the small number of normal lungs evaluated. The LLSIR could provide an objective noninvasive method for quantifying fetal lung development that can be applied to image analysis post hoc. A normal range for LLSIR could also enable future application of this scale in the assessment of fetuses at risk for pulmonary hypoplasia. The identification of fetuses with pulmonary hypoplasia would be of considerable clinical importance in the antenatal care of fetuses at risk for this condition. Standard prenatal screening for fetal anomalies by using ultrasonography (US) begins at 16 weeks in our referral base, with subsequent follow-up US performed as clinically indicated. Therefore, the purpose of this study was to define retrospectively a normal range for LLSIR in fetuses of 16–40 weeks gestation by using half-Fourier single-shot turbo spin-echo MR imaging.
|
MATERIALS AND METHODS
|
|---|
Patients
Between July 1999 and July 2003, 157 pregnant women were referred from the Northern Alberta Perinatal Program for MR imaging evaluation of suspected fetal or uterine anomalies. Approval from the regional ethics review board for retrospective evaluation of these patients was obtained, and informed consent was waived. Follow-up regarding the clinical outcome of all fetuses imaged was achieved through chart review and/or contact with the child’s physician by letter and/or telephone interview (L.J.B., R.B.). A normal population was defined as those fetuses in which clinical assessment showed normal pulmonary function after birth or normal lung morphologic and histologic features at autopsy. A total of 74 fetuses with normal lungs were identified in 73 of 157 pregnancies (71 singleton pregnancies and two dichorionic twin gestations, including one twin gestation in which one fetus died, and only the surviving fetus was available for analysis). The remaining 84 pregnancies were excluded because of fetal renal abnormality (n = 26), fetal lung and/or diaphragm abnormality (n = 25), fetal demise without autopsy (n = 8), fetal cardiac anomaly (n = 6), non-renal-associated oligohydramnios (n = 4), fetal liver abnormality (n = 3), fetus not born at time of analysis (n = 3), trisomy 18 syndrome (n = 2), MR images not available (n = 3), mother less that 18 years old (n = 2), fetal lungs not visible at MR imaging (n = 1), and fetus less than 16 weeks gestation (n = 1).
Among the 84 excluded pregnancies, 22 fetuses underwent autopsy subsequent to demise. Review of the autopsy reports identified 10 fetuses (all from singleton pregnancies) with hypoplastic lung(s) (L.J.B., R.B.). Pathologic criteria for hypoplastic lung diagnosis included a lung-to-body weight ratio of less than 0.015 for fetuses of less than 28 weeks gestation and a lung-to-body ratio of less than 0.012 for fetuses of 28 weeks gestation or more (11). On the basis of the classification criteria previously described (12), hypoplastic lungs were categorized as either associated with oligohydramnios or not associated with oligohydramnios because these two conditions are thought to be structurally different.
MR Imaging and Analysis
At the time of MR imaging, all mothers included in this study were older than 18 years of age, were carrying fetuses of 16 weeks gestation or more, were mentally capable of signing informed consent, and did not have any contraindication for MR imaging (eg, pacemaker). Gestational age at the time of MR imaging for fetuses with normal lungs was determined from the mother’s recollection of last menstrual period and was confirmed at US (n = 40) or was adjusted for last menstrual period at US (n = 33). Gestational age at the time of MR imaging for fetuses with hypoplastic lungs was determined from the mother’s recollection of last menstrual period and was confirmed at US (n = 7) or was adjusted for last menstrual period at US (n = 3) (R.S.C.). MR imaging was performed in all fetuses by using a 1.5-T imager (Symphony; Siemens, Erlangen, Germany) with a phased-array surface coil. Half-Fourier single-shot turbo spin-echo sequences (1100/68 [repetition time msec/echo time msec], 149° flip angle, 19 sections, 4- or 5-mm section thickness, one signal acquired, acquisition time of 21 seconds) that were oriented in the transverse, sagittal, and coronal planes relative to the fetal position were performed. No sedative was administered during imaging.
MR images were analyzed on an MR workstation (Syngo Leonardo; Siemens). Because images of only one lung were available for seven of the 74 fetuses, a total of 141 normal lungs were examined. Six of the fetuses were imaged multiple times at different gestational ages. Our analysis includes only the final antenatal imaging examination in these six fetuses so that results from only one examination were used for all 74 fetuses. Eighteen hypoplastic lungs were reviewed in 10 fetuses because one fetus with hypoplasia had only a single lung that was visible on MR images, and another fetus had unilateral lung hypoplasia at autopsy. MR images were interpreted by one author (R.B.) with 5 years experience reading MR images of the fetus. LLSIR calculations were performed by one author (L.J.B.) with less than 1 year of experience reading MR images of the fetus. A region-of-interest tool was placed within a homogeneous portion of lung and liver from the same series of images by using a method modified from one previously described (10). The portion of the lung and liver that was chosen was visibly free of intraparenchymal vessels and adjacent structures. The region-of-interest tool was placed without consideration of the distance from the receiver coil, thereby providing a quantitative value of signal intensity (Fig 1). The area of the tool ranged from 0.3 to 1.3 cm2. For each lung examined, three lung-to-liver ratio readings were used to obtain an average ratio. Each of the three ratios calculated per lung was taken from a different image and/or section and, when possible, from a different orientation plane.
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Coronal half-Fourier single-shot turbo spin-echo MR image (1100/68, 4-mm section thickness, 30 x 30-cm field of view, 218 x 256 matrix) of fetus at 28 weeks gestation shows regions of interest in lung (upper region of interest) and liver (lower region of interest) from which LLSIR was calculated. Areas chosen in each organ were homogeneous and free of organ borders and vascular structures.
|
|
Statistical Analysis
One-way repeated-measures analysis of variance F tests were used to assess differences among the three LLSIR values averaged from the same lung. Numerical summaries, such as mean and standard deviation, summarize the LLSIR and gestational ages. The correlation between the right and left lung of the same fetus was calculated. A paired t test was used to determine whether the difference in mean LLSIR between right and left lungs of the entire population was statistically significant. A mixed-effects model was developed (Y.L.) to capture the relationship between gestational age and LLSIR by using commercially available statistical software packages (SAS, version 8; SAS Institute, Cary, NC; Splus, Mathsoft, Seattle, Wash). A backward model building approach was used. A mixed model is a generalization of the standard linear model in the sense that predictors are entered into the model as an ordinary linear regression, but an additional term, called random effect, is also added. Because repeated measurements were taken in the same fetus, these measurements were correlated. A mixed model is needed to capture the dependence of the left and right lungs in the same fetus through a fetus random effect. The fetus random effect provides a way of modeling the mean, as well as variances and correlations, of the data. The mixed model can be calculated by using statistical software (SAS PROC MIXED; SAS Institute). A 95% prediction interval for normal lungs was determined on the basis of results from the mixed model. These prediction intervals provide limits for the prediction of a new LLSIR score for a new lung at a specified gestational age. A P value of less than .05 was considered to indicate a statistically significant difference.
|
RESULTS
|
|---|
LLSIR Measurements
The three left lung LLSIR measurements had means of 2.435 ± 0.765 (± standard deviation), 2.488 ± 0.850, and 2.366 ± 0.720. Right lung measurements had means of 2.357 ± 0.576, 2.390 ± 0.636, and 2.360 ± 0.684. No significant differences between the three averaged LLSIR measurements per lung were found in the left (F2,132 = 2.314, P = .103) or right (F2,146 = 0.280, P = .756) lungs. Averaging the repeated measure is appropriate given the lack of evidence of a significant difference. The LLSIR of normal lungs is illustrated according to gestational age in Figure 2. Fetuses with normal lung function had a mean gestational age of 28.0 weeks ± 5.4 (median age, 29 weeks; age range, 16–40 weeks). LLSIR was obtained for 74 right lungs and 67 left lungs. The left lungs had a mean LLSIR of 2.437 ± 0.745 (median, 2.16; range, 1.52–4.31), and the right lungs had a mean LLSIR of 2.369 ± 0.587 (median, 2.22; range, 1.57–3.89). The mean LLSIR for left lungs was not significantly different from the mean LLSIR for right lungs (paired t test, t = 1.799; df = 66; P = .077). The left and right lungs from the same fetus were highly linearly related, with a correlation of 0.73 (95% confidence intervals: 0.68–0.78). In the normal population, LLSIR varied from 1.52 at 21 weeks gestation to 4.31 at 34 weeks gestation. The LLSIR was higher in fetuses of more advanced gestational age.
View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Scatter plot of LLSIR for normal lungs. Mean curve (solid line) and 95% prediction interval (dotted lines) for each gestational age are shown and are based on mixed-effects statistical model. LLSIR increases with age, likely because of increased lung fluid retained within the lung as lung spaces develop.
|
|
Hypoplastic Lungs
Eighteen hypoplastic lungs were identified at autopsy in 10 fetuses. Information regarding these cases is summarized in Table 1, and the data are plotted in Figure 3. The mean gestational age of fetuses with hypoplastic lungs was 22.7 weeks ± 4.3; these fetuses had a mean LLSIR of 1.500 ± 0.180. Hypoplastic lungs associated with oligohydramnios had a mean signal intensity ratio of 1.370 ± 0.090 (range, 1.22–1.50), whereas hypoplastic lungs not associated with oligohydramnios had a mean signal intensity ratio of 1.660 ± 0.110 (range, 1.55–1.86). The LLSIR scores for hypoplastic lungs were usually less than the LLSIR scores for normal lungs in fetuses of similar gestational age.
Mixed-Model Results
To capture the relationship between LLSIR and gestational age, a mixed model was fit to adjust for the dependence of lung measurements from the same fetus (Table 2). The fitted model suggests a quadratic relationship between gestational age and LLSIR. Although the P value for the quadratic term in gestational age is greater than .05, the quadratic term is the most significant, and our approach dictates that lower order terms be included if higher order terms are in the model. The P value for the quadratic age term is greater than .05 because of the inclusion of age in the model. Left and/or right lung effect was not included in the final model because, when incorporated, the lung effect was not significant (P = .102). The model assumes that the correlation between the left and right lungs from the same fetus is the same for each fetus, and, with this assumption, the estimated correlation was 0.51. The within-fetus variance of 0.1368 estimates the dependence of lungs from the same fetus.
The 95% prediction intervals for a new fetus at a given gestational age can be calculated on the basis of the fitted model. Because the mixed-effect model allows for variability within subjects, multiple LLSIR predictions can be obtained at a specified gestational age. These intervals provide limits for the prediction of LLSIR in a new fetus at a specified gestational age. Figures 2 and 3 provide the estimated average LLSIR value at each gestational age for normal lungs, as well as the lower and upper 95% prediction intervals.
All the abnormal lungs we assessed in fetuses of 25 weeks gestation or more had LLSIR scores that were lower than the 95% prediction lower bound (Fig 3). When assessed at 18 weeks gestation, the lungs of fetuses with hypoplasia and associated oligohydramnios were significantly different from those of fetuses in the normal population. Several normal lungs exceeded the upper 95% prediction value.
|
DISCUSSION
|
|---|
Normal lungs that were assessed in this study showed a higher LLSIR at more advanced gestational ages. This finding agrees with previous reports of increasing signal intensity and relaxation times in fetal lungs on T2-weighted MR images through gestation (9,14). Fetal lung fluid is an essential component of lung development. It is likely that the amount of lung fluid that is present in the lungs contributes to the increase in signal intensity ratio with time. During the lung maturation process through the canalicular and alveolar stage, bronchiole lumina become larger, and the multiplication of respiratory bronchioles, terminal sacs, and alveoli occurs (15). From 29 weeks to term, alveoli multiply from 29 million to approximately 150 million, which is associated with an increase in surface area and decreasing interstitium thickness between alveoli (16). This maturation process increases the volume available for lung fluid. Accompanying this, the secretion rate of lung fluid increases with gestational age as a result of the developing pulmonary microvascular and epithelial surface area (17). Hypoplastic lungs have fewer and smaller peripheral airspaces and a reduced number of airway branches, arteries, and veins (18). This pathologic condition would result in reduced capacity to contain fetal lung fluid, which is likely to contribute to the lower lung-to-liver signal intensities that were seen in the cases of hypoplasia we evaluated. We did not evaluate normal lungs that exceeded the upper 95% prediction value because our interest focused on the lower limit as a means to separate normal lungs from hypoplastic lungs.
Our data for hypoplastic lungs are in agreement with the data of Kuwashima et al (10). Together, both sets of data suggest that the LLSIR remains low in hypoplastic lungs over time, thereby amplifying the difference between normal lungs and hypoplastic lungs at more advanced gestational ages. Our data suggest that there is a marked difference between normal lungs and hypoplastic lungs at 25 weeks gestation or more. The distinction between hypoplastic lungs and normal lungs at less than 25 weeks gestation is less clear. When considering the normal evolution of fetal lung development, it is possible that normal lungs and hypoplastic lungs share similarities prior to 25 weeks gestation.
Hypoplastic lungs have been divided into two major groups. Wigglesworth and Desai (12) reported that hypoplastic lungs associated with anything other than oligohydramnios have reduced cell number but are structurally normal for gestational age, whereas hypoplastic lungs associated with oligohydramnios are biochemically and structurally immature for gestational age, having reduced cell number, low lung phospholipid content, and impaired development of epithelial and interstitial tissues. Wigglesworth and Desai (12) also suggested that the structural maturation arrest seen in cases of hypoplasia with oligohydramnios is a consequence of a total failure of liquid retention in the lungs. This loss of liquid retention is likely the result of pressure effects and spinal flexion (19). In cases of lung hypoplasia without associated oligohydramnios, such as a thoracic volume reduction, liquid retention in the lungs could still occur (13). Considering our limited data, there may be a difference with regard to LLSIR between hypoplastic lungs associated with oligohydramnios and those that are not associated with oligohydramnios. At less than 25 weeks, three (50%) of six hypoplastic lungs associated with oligohydramnios were outside the 95% confidence interval. For hypoplastic lungs not associated with oligohydramnios in this age category, all six lungs were within normal LLSIR range. The structural variation between lungs associated with oligohydramnios and those that are not associated with oligohydramnios may explain this difference.
There are several limitations to consider with our study. First, the use of MR imaging as a screening tool for fetuses in the normal population is not currently feasible, and, therefore, we were dependent on the retrospective review of a patient population that was referred for suspected uterine or fetal anomalies. While we excluded all fetuses with abnormalities believed to have a potential influence on the normality of the lung, we cannot be completely certain that such exclusion occurred. Second, there is a limited amount data for normal lungs in some gestational age groups. Third, the LLSIR was not obtained at the same distance from the surface coil for each fetus. The signal drop-off associated with this distance may alter the ratios in an unknown manner.
Data from this study have derived a normal range with a 95% prediction interval for LLSIR that will serve as a reference for future studies and aid in the assessment of fetal lung appearance. Our work supports a potential role for MR imaging-derived LLSIR in the antenatal diagnosis of lung hypoplasia, especially after 25 weeks gestation. This information could assist prenatal counseling and optimize patient care. Future studies will include a prospective look at cases of pulmonary hypoplasia and correlate LLSIR with outcome morbidity and mortality.
|
ACKNOWLEDGMENTS
|
|---|
The authors thank B. F. (Peter) Mitchell, MD, Perinatal Research Centre, Edmonton, Alberta, Canada, for reviewing the manuscript.
|
FOOTNOTES
|
|---|
Abbreviations: LLSIR = lung-to-liver signal intensity ratio,
SD = standard deviation
Authors stated no financial relationship to disclose.
Author contributions: Guarantor of integrity of entire study, R.B.; study concepts, R.B., R.S.C.; study design, R.B., R.S.C., L.J.B.; literature research, L.J.B., R.B., R.S.C.; clinical studies, R.B., R.S.C.; data acquisition, L.J.B., R.B.; data analysis/interpretation, L.J.B., Y.L., R.B.; statistical analysis, Y.L.; manuscript preparation, L.J.B.; manuscript definition of intellectual content, all authors; manuscript editing, L.J.B., R.S.C., R.B.; manuscript revision/review and final version approval, all authors
|
REFERENCES
|
|---|
- Goldstein RB. Ultrasound evaluation of the fetal thorax. In: Callen PW, eds. Ulrasonography in obstetrics and gynecology. 4th ed. Philadelphia, Pa: Saunders, 2000; 429-432.
- Liu X, Ashtari M, Leonidas JC, Chan Y. Magnetic resonance imaging of the fetus in congenital intrathoracic disorders: preliminary observations. Pediatr Radiol 2001; 31:435-439.[CrossRef][Medline]
- Walsh DS, Hubbard AM, Olutoye OO, et al. Assessment of fetal lung volumes and liver herniation with magnetic resonance imaging in congenital diaphragmatic hernia. Am J Obstet Gynecol 2000; 183:1067- 1069.[CrossRef][Medline]
- Coakley FV, Lopoo JB, Lu Y, et al. Normal and hypoplastic fetal lungs: volumetric assessment with prenatal single-shot rapid acquisition with relaxation enhancement MR imaging. Radiology 2000; 216:107-111.[Abstract/Free Full Text]
- Rypens F, Metens T, Rocourt N, et al. Fetal lung volume: estimation at MR imaging—initial results. Radiology 2001; 219:236-241.[Abstract/Free Full Text]
- Paek BW, Coakley FV, Lu Y, et al. Congenital diaphragmatic hernia: prenatal evaluation with MR lung volumetry—preliminary experience. Radiology 2001; 220:63-67.[Abstract/Free Full Text]
- Mahieu-Caputo D, Sonigo P, Dommergues M, et al. Fetal lung volume measurement by magnetic resonance imaging in congenital diaphragmatic hernia. BJOG 2001; 108:863-868.[CrossRef][Medline]
- Wedegaertner U, Tchirikov M, Habermann C, et al. Fetal sheep with tracheal occlusion: monitoring lung development with MR imaging and B-mode US. Radiology 2004; 230:353-358.[Abstract/Free Full Text]
- Levine D, Barnewolt CE, Mehta TS, Trop I, Estroff J, Wong G. Fetal thoracic abnormalities: MR imaging. Radiology 2003; 228:379-388.[Abstract/Free Full Text]
- Kuwashima S, Nishimura G, Iimura F, et al. Low-intensity fetal lungs on MRI may suggest the diagnosis of pulmonary hypoplasia. Pediatr Radiol 2001; 31:669- 672.[CrossRef][Medline]
- Ikeda K, Hokuto I, Mori K, et al. Intrauterine MRI with single-shot fast-spin echo imaging showed different signal intensities in hypoplastic lungs. J Perinat Med 2000; 28:151-154.[CrossRef][Medline]
- Wigglesworth JS, Desai R. Use of DNA estimation for growth assessment in normal and hypoplastic fetal lungs. Arch Dis Child 1981; 56:601-605.[Abstract]
- Greenough A. Pulmonary hypoplasia. In: Greenough A, Milner AD, Robertson NRC, eds. Neonatal respiratory disorders. London, England: Arnold, 1996; 436-447.
- Duncan KR, Gowland PA, Freeman A, Moore R, Baker PN, Johnson IR. The changes in magnetic resonance properties of the fetal lungs: a first result and a potential tool for the non-invasive in utero demonstration of fetal lung maturation. Br J Obstet Gynaecol 1999; 106:122-125.[Medline]
- Moore KL. The respiratory system In: The developing human: clinically oriented embryology. 3rd ed. Philadelphia, Pa: Saunders, 1982; 219-226.
- Hislop A. Fetal and postnatal anatomical development. In: Greenough A, Milner AD, Robertson NRC, eds. Neonatal respiratory disorders. London, England: Arnold, 1996; 3-12.
- Greenough A. Lung maturation. In: Greenough A, Milner AD, Robertson NRC, eds. Neonatal respiratory disorders. London, England: Arnold, 1996; 13-26.
- Wigglesworth JS, Desai R, Guerrini P. Fetal lung hypoplasia: biochemical and structural variations and their possible significance. Arch Dis Child 1981; 56:606-615.[Abstract]
- Albuquerque CA, Smith KR, Saywers TE, Johnson C, Cock ML, Harding R. Relation between oligohydramnios and spinal flexion in the human fetus. Early Hum Dev 2002; 68:119-126.[CrossRef][Medline]
TOP
ABSTRACT
INTRODUCTION
= hypoplastic lungs associated with oligohydramnios. + = hypoplastic lungs not associated with oligohydramnios. Data on hypoplastic lungs from Kuwashima et al (
) are included for comparison. At 25 weeks and beyond, all LLSIRs for hypoplastic lungs are outside the lower bound.