- Email: [email protected]
Ultrasonographic prediction of lethal pulmonary hypoplasia: Comparison of eight different ultrasonographic parameters
Ultrasonographic prediction of lethal pulmonary hypoplasia: Comparison of eight different ultrasonographic parameters Shuichiro Yoshimura, MD, Hideaki Masuzaki, MD, Hideo Gotoh, MD, Hisanobu Fukuda, MD, and Tadayuki Ishimaru, MD, PhD Nagasaki, Japan OBJECTIVE: The aim of this study was to determine the usefulness of eight different ultrasonographic fetal parameters for predicting fetal pulmonary hypoplasia. STUDY DESIGN: Nomograms of eight different ultrasonographic fetal parameters were evaluated by studying uncomplicated single fetus pregnancies with well-established dates between 18 and 40 weeks of gestation. The eight parameters, which could reflect fetal lung mass, were as follows: thoracic circumference, thoracic area, thoracic area minus heart area, lung area, thoracic circumference/abdominal circumference ratio, thoracic area/heart area ratio, thoracic area minus heart area/thoracic area ratio, and lung area/thoracic area ratio. The relative efficacy of the eight parameters was determined by studying 21 fetuses at high risk for development of lethal pulmonary hypoplasia and 30 fetuses with premature rupture of membranes within 1 week. RESULTS: The lung area (gestational age-dependent parameter) and the thoracic circumference/abdominal circumference (gestational age-independent parameter) ratio had the best diagnostic accuracy (sensitivity 81.3% and 90.5%, specificity 100% and 90.0%, positive predictive value 100% and 86.4%, negative predictive value 90.9% and 93.1%, respectively). There were significant linear relationships between lung weight and lung area and between the lung weight/body weight ratio and the thoracic circumference/abdominal circumference ratio. CONCLUSION: These data suggested that the application of lung area and the thoracic circumference/abdominal circumference ratio are clinically useful for the evaluation of fetal pulmonary hypoplasia. (Am J Obstet Gynecol 1996;175:477-83.)
Key words: Ultrason0graphy, fetal pulmonary hypoplasia, prenatal diagnosis
The presence of fetal pulmonary hypoplasia has been well documented in prolonged severe oligohydramnios caused by bilateral renal agenesis, infantile polycystic kidneys, posterior urethral valve, and prolonged leakage of amniotic fluid)' 2 Nimrod et al? reported a 9% incidence of lung hypoplasia after very prolonged rupture of the membranes. From the clinician's standpoint, the ability to accurately predict the occurrence of antepartum lethal pulmonary hypoplasia would be beneficial in both obstetric management and parental counseling. In the past several ultrasonographic techniques have been reported for the diagnosis of pulmonary hypoplasia: chest circumference, 4 chest area, chest area minus heart area, chest circumference/abdominal circumference ratio,5. 6 chest a r e a / h e a r t area ratio, and chest area minus heart area divided by chest area ratio7 In a previous study8
From the Department of Obstetrics and Gynecology,Nagasaki University School of Medicine. Received for publication June 19, 1995; revised October 17, 1995; accepted March 14, 1996. Reprint requests: Shuichiro Yoshimura, MD, Department of Obstetrics and Gynecology, Nagasaki University School of Medicine, 1-7-1 Sakamoto-machi, Nagasaki, 852, Japan. Copyright © 1996 by Mosby-YearBook, Inc. 0002-93 78/96 $5. O0+ 0 6/1/73874
we reported eight different ultrasonographic parameters in both uncomplicated pregnancies and pregnancies complicated by preterm premature rupture of membranes and suggested that these measurements were helpful in predicting lethal pulmonary hypoplasia. The objectives of the current study were (1) to establish nomograms of the eight different ultrasonographic parameters, (2) to assess the effect of premature rupture of membranes on these parameters and to determine the usefulness of the parameters in predicting lethal pulmonary hypoplasia in high-risk fetuses, and (3) to examine the relationships between each parameter and lung weight or lung weight/body weight ratio. Material and methods
In the current study the first step was to establish the nomograms of the eight fetal parameters throughout gestation, A total of 150 uncomplicated pregnancies between 18 and 40 weeks of gestation (dates confirmed by early ultrasonography) were used for single ultrasonographic measurement to generate the nomograms in a cross-sectional manner. All patients had intact membranes and singleton pregnancies and were delivered by spontaneous labor or elective repeat cesarean section. 477
478 Yoshimuraet al.
August 1996 AmJ ObstetGynecol
Fig. 1. Uhrasonographic photograph showing cross-section of fetal chest at level of four-chamber view, illustrating placement of calipers for measurements of heart area and lung area. Areas are automatically calculated by counting numbers of pixels in area with computer system. Lung area corresponded with Area (X) -area (+).
Patients with fetal abnormalities (e.g., congenital anomalies, intrauterine growth restriction, or fetal macrosomia) were excluded. The uhrasonographic examinations were performed with a linear-array real-time ultrasonographic system (Yokogawa [Tokyo] RT8000 or RT2800) with a 3.5 MHz transducer. Freeze-frame capabilities were available, and on-screen calipers were used for the measurements. In addition to routine biometry, the bony thoracic circumference, thoracic area, heart area, and lung area were determined from a cross-section of the fetal chest at the four-chamber view level with the heart in ventricular diastole (Fig. 1). From this view the thoracic circumference was calculated by the anteroposterior diameter and transverse diameter, and the thoracic area heart area and lung area were automatically calculated by counting numbers of pixels in the area with the computer system. The relationship between gestational age and each of the following eight parameters was defined: thoracic circumference, thoracic area, thoracic area minus heart area, lung area, thoracic circumference/abdominal circumference ratio, thoracic area/heart area ratio, thoracic area minus heart area/thoracic area ratio, and lung area/thoracic area ratio. Scattergrams with increasing gestational age were developed, and regression analyses were used to establish the nomograms, including - 2 SD, mean, and +2 SD lines. The statistical analysis of the data was performed by computer (Macintosh [Apple, Cupertino, Calif.] LC630) with StatView (Berke-
ley, Calif.) and DeltaGraph (Monterey, Calif.) software. After the nomograms were established, the second step was to assess the efficacy of each parameter by studying 30 patients with ruptured membranes not at risk for pulmonary hypoplasia (duration of premature rupture of membranes within 1 week) and 21 patients at risk for pulmonary hypoplasia. The gestational age at premature rupture of membranes in 30 patients was 27 weeks (4 cases), 28 weeks (4 cases), 29 weeks (4 cases), 30 weeks (3 cases), 31 weeks (3 cases), 32 weeks (4 cases), 33 weeks (4 cases), and 34 weeks (4 cases). All 30 patients with premature rupture of membranes underwent a single measurement within 1 week of delivery, and none of the infants had any sign of pulmonary hypoplasia. Thirteen of the 21 at risk for pulmonary hypoplasia had renal anomalies (5 cases of bilateral renal agenesis, 5 of unilateral muhicystic kidney with contralateral renal agenesis, and 3 infantile polycystic kidney disease), 6 had thanatophoric dwarfism, and 2 had premature rupture of membranes of >5 weeks' duration at <26 weeks of gestation. The last examination within 1 week of delivery was used for analysis. In all 21 cases thoracic circumference, thoracic area, and thoracic circumference/abdominal circumference ratio were measured, and thoracic area minus heart area, lung area, thoracic area/heart area ratio, thoracic area minus heart area/thoracic area ratio, and lung area/thoracic area ratio were measured in 16 of 21. All 21 high-risk fetuses had lethal pulmonary hypoplasia (Table I). Measurements <2 SD were considered abnormal. All neonates
Volume 175, Number 2 AmJ ObstetGynecol
Yoshimura et al.
479
Table I. Summary of clinical features of lethal fetal pulmonary hypoplasia Gestational age at delivery (wk)
Birth weight (gin)
1 2 3 4 5 6 7 8
21 29 34 35 36 31 34 20
430 1150 1770 2014 800 1650 2540 480
9
25
830
10
31
1700
11
32
1670
12
34
2540
13
37
1880
14 15 16 17 18 19 20 21
21 32 30 31 31 31 39 39
600 1795 1110 1698 1850 1850 2160 2960
Case No.
Diagnosis
Bilateral renal agenesis Bilateral renal agenesis Bilateral renal agenesis Bilateral renal agenesis Bilateral renal agenesis Infantile polycystickidney Infantile polycystickidney Unilateral renal agenesis and contralateral nmlticystic kidney Unilateral renal agenesis and contralateral multicystic kidney Unilateral renal agenesis and contralateral muldcystic kidney Unilateral renal agenesis and contralateral multicystic kidney Unilateral renal agenesis and contralateral multicystic kidney Unilateral renal agenesis and contralateral multicystic kidney Preterm PROM (>5 wk) Preterm PROM (>8 wk) Thanatophoric dwarfism Thanatophoric dwarfism Thanatophoric dwarfism Thanatophoric dwarfism Thanatophoric dwarfism Thanatophoric dwarfism
Outcome
Stillbirth Neonatal Neonatal Stillbirth Neonatal Neonatal Neonatal Neonatal
death death death death death death
Stillbirth Neonatal death Neonatal death Neonatal death Neonatal death Stillbirth Neonatal death Neonatal death Neonatal death Neonatal death Neonatal death Neonatal death Neonatal death
PROM, Premature rupture of membranes.
with lethal pulmonary hypoplasia died in the first 24 hours after birth, and the diagnosis was confirmed at autopsy by lung weight/body weight ratios and radial alveolar counts. 9 The measurements of the groups with preterm premature rupture of membranes versus those with pulmonary hypoplasia were compared by Fisher's exact and two-tailed t tests. A value ofp < 0.01 was considered statistically significant. To determine the relationship between each parameter and lung weight or lung weight/body weight ratio, three patients (anhydramnios without pulmonary hypoplasia, hydrocephalus, and Turner syndrome) who could undergo measurements for the eight parameters were included, and regression analyses were performed. Results There was a significant linear relationship between thoracic circumference and gestational age (y -- 0.834x2.477 [R2= 0.948, p < 0.001]), between thoracic area and gestational age (y= 2.846x- 42.54 [R2=0.929, p < 0.001]), between thoracic area minus heart area and gesrational age (y= 2.128x- 31.29 JR2= 0.906, p < 0.001]), and between lung area and gestational age (y = 1.074x15.78 [R2= 0.866, p < 0.001]). Regression analyses of the relationships among the four ratios (thoracic circumference/abdominal circumference, thoracic area/heart area, thoracic area minus heart area/thoracic area, and lung area/thoracic area)
and gestational age revealed these ratios to be relatively constant. The mean value was 0.888 (SD 0.044) for the thoracic circumference/abdominal circumference ratio, 4.05 (SD 0.51) for the thoracic area/heart area ratio, 0.744 (SD 0.034) for the thoracic area minus heart area/thoracic area ratio, and 0.403 (SD 0.052) for the lung area/thoracic area ratio. After the establishment of nomograms, the efficacy of parameters was assessed by studying 30 patients with premature rupture of membranes (gestational ages 27 to 34 weeks) and 21 patients with pulmonary hypoplasia (gestational ages 20 to 39 weeks). The measurements for these patients were compared with the nomograms (-2 SD lines) established from the patients with intact membranes. Thoracic circumference was at or below the - 2 SD line in 5 of the 30 patients with premature rupture of membranes (16.7%) and in all 21 with pulmonary hypoplasia (100%) (p < 0.01). Thoracic area was at or below the - 2 SD line in 4 of the 30 with premature rupture of membranes (3.3%) but in all 21 with pulmonary hypOplasia (100%) (p < 0.01). Thoracic area minus heart area was at or below the - 2 SD line in 4 of the 30 with premature rupture of membranes (13.3%) and in all 6 with pulmonary hypoplasia (100%) (p < 0.01), and lung area was at or below the - 2 SD line in none of the 30 with premature rupture of membranes (0%) but in 13 of the 16 with pulmonary hypoplasia (81.3%) (p < 0.01) (Figs. 2 and 3). The thoracic circumference/abdominal circum-
480
Yoshimura et al.
August 1996
AmJ Obstet Gynecol
35"
9080-
30"
70-
~ - 2 S D
25-
0 o 8
oo°
20-
60" .
o. ...
20i0-
10
0 20
25
30
35
40
/
. 8" u 20
gestational age (weeks) 7060-
•
/ °o
20-
10- o._~-----"~•
°
-2SD
8 e •
.• .
•
e iw
i u 30 35 gestationat age (weeks)
i
40
/
30-
o o ~ o o
•
i 25
35-
= 4o~o
D
o'l
40-
u~ 50 <
30-
S
30"
• ..
15"
~
50" 40"
E 25< .n 2015-
•
I0-
•
.
50
20 '
2'5 30 ' 35 ' gestational age (weeks)
40 '
20
25 30 gestational age
35
(weeks)
40
Fig. 2. Distribution of thoracic circumference (TC), thoracic area (TA), thoracic area minus heart area (HA), and lung area (LA) in preterm premature rupture of membranes and pulmonary hypoplasia. Open circles,Preterm premature rupture of membranes; solid circles,pulmonary hypoplasia. Values <-2 SD were considered abnormal.
Table II. Mean _+ SD of gestational a g e - i n d e p e n d e n t parameters in three groups
Control Preterm PROM Pulmonary hypoplasia
TC/AC
TA/HA
TA- HA/TA
LA/TA
0.888 + 0.044* 0.887 - 0.054* 0.748 _ 0.076*
4.05 -+ 0.51 t 4.05 + 0.48]3.15 + 0.84]-
0.744 + 0.0345 0.742 _+0.026 + 0.664 +_0.076 +
0.403 -+ 0.052 0.420 -+ 0.079 0.351 + 0.096
TC, Thoracic circumference; AC, abdominal circumference; TA, thoracic area; HA, heart area; LA, lung area; PROM, premature rupture of membranes. *Control and preterm PROM versus pulmonary hypoplasia, p < 0.01. tControl and preterm PROM versus pulmonary hypoplasia, p < 0.01. $Control and preterm PROM versus pulmonary hypoplasia, p < 0.01.
ference ratio was 0.887_+ 0.054 in preterm premature rupture of m e m b r a n e s but 0.748-+ 0.076 (p< 0.01) in p u l m o n a r y hypoplasia, the thoracic area/heart area ratio was 4.05 + 0.048 in preterm premature rupture of m e m branes but 3.15 + 0.084 in p u l m o n a r y hypoplasia (p< 0.01), the thoracic area minus heart area/thoracic area ratio was 0.742 _+ 0.026 in preterm premature rupture of m e m b r a n e s but 0.664 + 0.079 in p u l m o n a r y hypoplasia ( p < 0 . 0 1 ) , and the lung area/thoracic area ratio was 0.420_+ 0.079 in preterm premature rupture of m e m branes but 0.351_+0.096 in p u l m o n a r y hypoplasia (Table II). The diagnostic index values of the eight ultrasonographic parameters for prediction of lethal p u l m o n a r y hypoplasia are shown in Table III. Lung area (gestational
a g e - d e p e n d e n t parameter) and thoracic circumfere n c e / a b d o m i n a l circumference ratio (gestational agei n d e p e n d e n t parameter) exhibited the highest diagnostic accuracy (sensitivity 81.3% and 90.5%, specificity 100% and 90.0%, positive predictive value 100% and 86.4%, and negative predictive value 90.9% and 93.1%, respectively). There was a significant linear relationship between lung weight and the gestational a g e - d e p e n d e n t parameters (thoracic circumference [y= 2 . 0 5 x - 20.50, / U = 0.613], thoracic area [y = 0 . 0 7 4 x - 2 . 9 8 , / ~ = 0.568], thoracic area m i n u s heart area [y = 0 . 1 x - 3.152, R 2 = 0.731], and lung area [y= 0 . 2 4 x - 6.021, / ~ = 0.786]) (Fig. 4). There was also a significant linear relationship between the lung w e i g h t / b o d y weight ratio and the gestational
Yoshimura et al.
Volume 175, N u m b e r 2 A m J O b s t e t Gynecol
1.1-
7-
1"
.£
88o80
~o
0.9"
o 0.8-
-w;----, "
00
6-
"+2SD
o
o 5-
°800
~
~
~
o o o ° O°•o~g o• O0
2i
3i0 35 gestati0nal age (weeks)
•
e
0.6-
2'5
~
24-
2sD
i
~
o
3-
20
/
•
0.7-
0.5
481
i
410
20
i
+2SD o o
o
i
*-
i
i
25 30 35 gestational age (weeks)
40
0.6"
+2SD 0.5-
.£ 0.9" .2
0.4" < E~ 0.3-
+2SD
0.8• o
@
o.7-
• •
°
•
•
• •
L.-----2SD
"---~-2sD
0.2"
0.60.5
oooo •
0.1-
%
i
25
i
30
~'s
0
i
40
gestationaI age {weeks)
2'o
~5
,'0
&
i
40
gestational age (weeks)
Fig. 3. Distribution of thoracic circumference (TC)/abdominal circumference (AC), thoracic area (TA)/heart area (HA), thoracic area minus heart area/thoracic area, and lung area (LA)/thoracic area ratio in preterm premature rupture of membranes and pulmonary hypoplasia. Open circles, Preterm premature rupture of membranes; solid circles, pulmonary hypoplasia. Values <-2 SD were considered abnormal. Table III. Diagnostic i n d e x values of lethal p u l m o n a r y hypoplasia
Gestational age dependent TC (n = 51) TA (n = 51) TA - HA (n = 46) LA (n = 46) Gestational age independent TC/AC (n = 51) TA/HA (n = 46) T A - HA/TA (n = 46) LA/TA (n = 46)
Sensitivity (%)
Specificity(%)
PPV (%)
NPV (%)
Accuracy (%)
100 100 100 81.3
83.3 86.7 86.7 100
80.8 84.0 80.0 100
100 100 100 90.9
90.2 92.2 91.3 93.5
90.5 68.8 68.8 31.3
90.0 100 96.7 100
86.4 100 91.7 100
" 93.1 85.7 85.3 73.2
90.2 89.1 87.0 76.1
PPV, Positive predictive value; NPV, negative predictive value; °o-. age-independent parameters (thoracic circumfere n c e / a b d o m i n a l c i r c u m f e r e n c e ratio [y = 0 . 0 7 8 x - 0.049, R 2 = 0.651], t h o r a c i c a r e a / h e a r t a r e a ratio [y = 0 . 0 0 4 9 x 0 . 0 0 6 2 , / ~ = 0.596], t h o r a c i c area m i n u s h e a r t a r e a / t h o racic a r e a ratio [y = 0 . 0 4 8 x - 0.022, R 2 = 0.40], a n d l u n g a r e a / t h o r a c i c area ratio [y=O.O295x-O.O012, R~= 0.186]) (Fig. 5). Comment
T h e p r e s e n c e of p u l m o n a r y hypoplasia has b e e n well d o c u m e n t e d in fetuses with l o n g - s t a n d i n g o l i g o h y d r a m nios, w h i c h c o u l d b e t h e r e s u l t o f c h r o n i c a m n i o t i c fluid leakage o r fetal r e n a l disease or it c o u l d b e idiopathic.
T h e e x t e n t o f fetal p u l m o n a r y hypoplasia d e p e n d s o n t h e degree and the duration of oligohydramnios and on the gestational age a n d stage o f l u n g d e v e l o p m e n t w h e n it occurs. F r o m t h e clinician's s t a n d p o i n t , it is desirable to b e able to distinguish hypoplasia as e i t h e r l e t h a l or n o n lethal. T h e c u r r e n t study was u n d e r t a k e n to clarify t h e clinical diagnosis o f p u l m o n a r y hypoplasia i n surviving infants, with t h e p r e d i c t i o n o f t h e l e t h a l f o r m o f p u l m o n a r y hypoplasia u s e d as a n e n d point. Previous investigators have a t t e m p t e d to achieve t h e a n t e n a t a l diagnosis of the lethal a n d n o n l e t h a l f o r m s o f p u l m o n a r y hypoplasia by use of n o m o g r a m s o f c h e s t c i r c u m f e r e n c e m e a s u r e m e n t s t h r o u g h o u t gestation 4 or
482
Yoshimura et al.
August 1996 AmJ Obstet Gynecol
40 n=10 r=0.89(p<0.001) y=2.41x-6.02
30
O
bJ0 O
20
ooo
10
0
•
•
4
6
I
I
I
I
I
8
10
12
14
16
LA(cm2) Fig. 4. Relationship between lung area (LA) and lung weight (LW).
0.030
0 n=17 0.025
0
iza
0.020 0.015
OO ~ ~ OO
0.010 0.005
O
/;9/°
0 0.65
°
O
I
I
I
I
I
I
0.70
0•75
0.80
0.85
0.90
0.95
T C / A C ratio Fig. 5. Relationship between thoracic circumference (TC)/abdominal circumference (AC) ratio and lung weight (LW)/body weight (BW) ratio. the chest c i r c u m f e r e n c e / a b d o m i n a l c i r c u m f e r e n c e ratio 5, 6 and the chest area minus heart a r e a / c h e s t area ratio, 7 which have the advantage of being constant t h r o u g h o u t gestation. T h e use of chest circumference, however, requires accurate knowledge of the length of gestation. T h e r e f o r e these n o m o g r a m s c a n n o t be used in fetuses with unknown gestational age or fetuses with suspected intrauterine growth restriction, which is frequently present• T h e chest c i r c u m f e r e n c e / a b d o m i n a l circumference ratio is gestational a g e - i n d e p e n d e n t and can be used reliably without accurate knowledge of the gestational age. T h e major limitation of this m e t h o d is that it c a n n o t be used in fetuses with large chest circumferences caused by the
excessive skin and subcutaneous tissue in disorders that affect or distort a b d o m i n a l c i r c u m f e r e n c e m e a s u r e m e n t s (e,g., fetuses with polycystic kidneys, obstructive uropathy, or omphalocele). It is known t h a t many cases of p u l m o n a r y hypoplasia are due to p r o l o n g e d r u p t u r e of membranes. Thus the use of the two most c o m m o n parameters (chest circumference and chest circumfere n c e / a b d o m i n a l circumference ratio) is questionable because the effect of p r e m a t u r e r u p t u r e of m e m b r a n e s on these m e a s u r e m e n t s has n o t b e e n investigated. With the above considerations in mind, we a t t e m p t e d in this cross-sectional study to establish n o m o g r a m s of ultrasonographic parameters that could possibly reflect fetal lung mass by studying 150 u n c o m p l i c a t e d pregnan-
Volume 175, Number 2 AmJ Obstet Gynecol
cies with intact membranes at the examination. The relationships between gestational age and thoracic circumference, thoracic area, thoracic area minus heart area, and lung area showed significant differences, but the four ratios (thoracic circumference/abdominal circumference, thoracic a r e a / h e a r t area, thoracic area minus heart area/thoracic area, and lung area/thoracic) were relatively constant. The chest circumference/abdominal circumference ratio was found to differ slightly from previously reported norms (mean value 0.91 to 0.94 in previous studies vs 0.88 in the current study). This difference can be explained by the fact that we measured the bony chest circumference (the method of Vintzileos et al. 7) instead of measuring from the outer edge to the outer edge of the fetal thorax as done previously. Although we used the method of Vintzilens et al., our results for the thoracic circumference/abdominal circumference ratio, the thoracic a r e a / h e a r t area ratio, and the thoracic area minus heart a r e a /t h o racic area ratio were somewhat higher than their norms. This difference may be due to the difference in body structure between Japanese and American fetuses. In patients with premature rupture of membranes the ruptured membrane itself is known to cause fetal compression and therefore may be inherently associated with alterations in fetal biometry. The current study addressed this point by evaluating fetuses with ruptured membranes who were unlikely to have pulmonary hypoplasia because the duration of premature membrane rupture was <1 week. In all cases with premature rupture of membranes the measurements were within the normal range for all three gestational a g e - i n d e p e n d e n t ratios. The fact that the - 2 SD line of the four ratios was not altered by the presence of premature rupture of membranes makes the ratios potentially valuable tools for evaluating fetuses with prolonged premature rupture of membranes or uncertain gestational age. As noted in Table III, most parameters had positive and negative predictive values that were too high to be clinically useful. The diagnostic accuracy of the studied parameters was much higher than that of previously reported parameters. 7 This discrepancy may be due to (1) differences in the populations studied, (2) inclusion of fetuses at particularly high risk for development of pulmonary hypoplasia in this study (all had documented oligohydramnios of >5 weeks' duration), (3) differences regarding the end point of prediction, which in our study was the lethal form of the disease, and (4) the difference in body structure between Japanese and American fetuses. In the current study the gestational development parameter lung area showed a significant linear relationship with lung weight, and the thoracic circumference/abdominal circumference ratio was significantly related to
Yoshimura et al. 483
the lung weight/body weight ratio. These relationships suggested that the thoracic circumference/abdominal circumference ratio was the best clinical parameter for the prediction of pulmonary hypoplasia. Recently several new techniques for the prediction of pulmonary hypoplasia have been reported, including fetal breathing movement, I° ductus arteriosus flow velocity modulation, 1~ and fetal upper respiratory tract function, I2 by Doppler ultrasonography. These methods have not been firmly established and thus require further investigation. In conclusion, lung area and the thoracic circumfere n c e / a b d o m i n a l circumference ratio had good diagnostic accuracy (93.5% and 90.3%) for antenatal diagnosis of pulmonary hypoplasia. This ratio shows significant linear relationships with lung weight (lung area) and the lung weight/body weight ratio (thoracic circumference/abdominal circumference ratio). These data suggested that the application of lung area and the thoracic circumfere n c e / a b d o m i n a l circumference ratio are clinically useful for the evaluation of fetal pulmonary hypoplasia. REFERENCES
1. Potter EL. Bilateral renal agenesis. J Pediatr 1946;29:68-73. 2. Thomas IT, Smith DW. Oligohydramnios, cause of the nonrenal features of Potter's syndrome, including pulmonary hypoplasia. J Pediatr 1974;84:811-4. 3. Nimrod C, Varela Gittings F, Machin G, Campbell D, Wesenberg R. The effect of very prolonged membrane rupture on fetal development. AmJ Obstet Gynecol 1984;148:540-4. 4. Nimrod C, Davies D, Iwanicki S, Harder J, Persaud D, Nicholson S. Ultrasound prediction of pulmonary hypoplasia. Obstet Gynecol 1986;68:496-8. 5. Fong K, Ohlsson A, Zalev A. Fetal thoracic circumference: a prospective cross-sectional study with real-time ultrasound. AmJ Obstet Gynecol 1988;158:1154-60. 6. Johnson A, Callan N, Bhutani VK, Colmorgen GHC, Weiner S, Bolognese RJ. Ultrasonic ratio of fetal thoracic to abdominal circumference: an association with fetal pulmonary hypoplasia. AmJ Obstet Gynecol 1987;157:764-9. 7. Vintzileos AM, Campbell WA, Rodis JF, Nochimson DJ, Pinette MG, Petrikovsky BM. Comparison of six different ultrasonographic methods for predicting lethal fetal pulmonary hypoplasia. AmJ Obstet Gynecol 1989;161:606-12. 8. Yoshimura S, Masuzaki H, Fukuda H, Yamabe T. Comparison of eight different ultrasonographic parameters for predicting lethal fetal pulmonary hypoplasia. Jpn J Med Ultrasonics 1992;19:207-13. 9. Askenazi SS, Perlmma M. Pulmonary hypoplasia: lung weight and radial count as criteria of diagnosis. Arch Dis Child 1979;54:614-8. 10. Blott M, Greenough A, Nicolaides KH, Moscoso G, Gibb D, Campbell S. Fetal breathing movements as predictor "of favourable pregnancy outcome after oligohydramnios due to membrane rupture in second trimester. Lancet 1987;1: 129-31. 11. EyckJV, Mooren K, WladimiroffJW. Ductus arteriosus flow velocity modulation by fetal breathing movements as a measure of fetal lung development. Am J Obstet Gynecol 1990; 163:558-66. 12. Fox HE, Badalian SS, Timor-Tritsch TE, Stolar CJH. Fetal upper respiratory tract function in cases of antenatally diagnosed congenital diaphragmatic hernia: preliminary observations. Ultrasound Obstet Gynecol 1993;3:164-7.
Key words: Ultrason0graphy, fetal pulmonary hypoplasia, prenatal diagnosis
The presence of fetal pulmonary hypoplasia has been well documented in prolonged severe oligohydramnios caused by bilateral renal agenesis, infantile polycystic kidneys, posterior urethral valve, and prolonged leakage of amniotic fluid)' 2 Nimrod et al? reported a 9% incidence of lung hypoplasia after very prolonged rupture of the membranes. From the clinician's standpoint, the ability to accurately predict the occurrence of antepartum lethal pulmonary hypoplasia would be beneficial in both obstetric management and parental counseling. In the past several ultrasonographic techniques have been reported for the diagnosis of pulmonary hypoplasia: chest circumference, 4 chest area, chest area minus heart area, chest circumference/abdominal circumference ratio,5. 6 chest a r e a / h e a r t area ratio, and chest area minus heart area divided by chest area ratio7 In a previous study8
From the Department of Obstetrics and Gynecology,Nagasaki University School of Medicine. Received for publication June 19, 1995; revised October 17, 1995; accepted March 14, 1996. Reprint requests: Shuichiro Yoshimura, MD, Department of Obstetrics and Gynecology, Nagasaki University School of Medicine, 1-7-1 Sakamoto-machi, Nagasaki, 852, Japan. Copyright © 1996 by Mosby-YearBook, Inc. 0002-93 78/96 $5. O0+ 0 6/1/73874
we reported eight different ultrasonographic parameters in both uncomplicated pregnancies and pregnancies complicated by preterm premature rupture of membranes and suggested that these measurements were helpful in predicting lethal pulmonary hypoplasia. The objectives of the current study were (1) to establish nomograms of the eight different ultrasonographic parameters, (2) to assess the effect of premature rupture of membranes on these parameters and to determine the usefulness of the parameters in predicting lethal pulmonary hypoplasia in high-risk fetuses, and (3) to examine the relationships between each parameter and lung weight or lung weight/body weight ratio. Material and methods
In the current study the first step was to establish the nomograms of the eight fetal parameters throughout gestation, A total of 150 uncomplicated pregnancies between 18 and 40 weeks of gestation (dates confirmed by early ultrasonography) were used for single ultrasonographic measurement to generate the nomograms in a cross-sectional manner. All patients had intact membranes and singleton pregnancies and were delivered by spontaneous labor or elective repeat cesarean section. 477
478 Yoshimuraet al.
August 1996 AmJ ObstetGynecol
Fig. 1. Uhrasonographic photograph showing cross-section of fetal chest at level of four-chamber view, illustrating placement of calipers for measurements of heart area and lung area. Areas are automatically calculated by counting numbers of pixels in area with computer system. Lung area corresponded with Area (X) -area (+).
Patients with fetal abnormalities (e.g., congenital anomalies, intrauterine growth restriction, or fetal macrosomia) were excluded. The uhrasonographic examinations were performed with a linear-array real-time ultrasonographic system (Yokogawa [Tokyo] RT8000 or RT2800) with a 3.5 MHz transducer. Freeze-frame capabilities were available, and on-screen calipers were used for the measurements. In addition to routine biometry, the bony thoracic circumference, thoracic area, heart area, and lung area were determined from a cross-section of the fetal chest at the four-chamber view level with the heart in ventricular diastole (Fig. 1). From this view the thoracic circumference was calculated by the anteroposterior diameter and transverse diameter, and the thoracic area heart area and lung area were automatically calculated by counting numbers of pixels in the area with the computer system. The relationship between gestational age and each of the following eight parameters was defined: thoracic circumference, thoracic area, thoracic area minus heart area, lung area, thoracic circumference/abdominal circumference ratio, thoracic area/heart area ratio, thoracic area minus heart area/thoracic area ratio, and lung area/thoracic area ratio. Scattergrams with increasing gestational age were developed, and regression analyses were used to establish the nomograms, including - 2 SD, mean, and +2 SD lines. The statistical analysis of the data was performed by computer (Macintosh [Apple, Cupertino, Calif.] LC630) with StatView (Berke-
ley, Calif.) and DeltaGraph (Monterey, Calif.) software. After the nomograms were established, the second step was to assess the efficacy of each parameter by studying 30 patients with ruptured membranes not at risk for pulmonary hypoplasia (duration of premature rupture of membranes within 1 week) and 21 patients at risk for pulmonary hypoplasia. The gestational age at premature rupture of membranes in 30 patients was 27 weeks (4 cases), 28 weeks (4 cases), 29 weeks (4 cases), 30 weeks (3 cases), 31 weeks (3 cases), 32 weeks (4 cases), 33 weeks (4 cases), and 34 weeks (4 cases). All 30 patients with premature rupture of membranes underwent a single measurement within 1 week of delivery, and none of the infants had any sign of pulmonary hypoplasia. Thirteen of the 21 at risk for pulmonary hypoplasia had renal anomalies (5 cases of bilateral renal agenesis, 5 of unilateral muhicystic kidney with contralateral renal agenesis, and 3 infantile polycystic kidney disease), 6 had thanatophoric dwarfism, and 2 had premature rupture of membranes of >5 weeks' duration at <26 weeks of gestation. The last examination within 1 week of delivery was used for analysis. In all 21 cases thoracic circumference, thoracic area, and thoracic circumference/abdominal circumference ratio were measured, and thoracic area minus heart area, lung area, thoracic area/heart area ratio, thoracic area minus heart area/thoracic area ratio, and lung area/thoracic area ratio were measured in 16 of 21. All 21 high-risk fetuses had lethal pulmonary hypoplasia (Table I). Measurements <2 SD were considered abnormal. All neonates
Volume 175, Number 2 AmJ ObstetGynecol
Yoshimura et al.
479
Table I. Summary of clinical features of lethal fetal pulmonary hypoplasia Gestational age at delivery (wk)
Birth weight (gin)
1 2 3 4 5 6 7 8
21 29 34 35 36 31 34 20
430 1150 1770 2014 800 1650 2540 480
9
25
830
10
31
1700
11
32
1670
12
34
2540
13
37
1880
14 15 16 17 18 19 20 21
21 32 30 31 31 31 39 39
600 1795 1110 1698 1850 1850 2160 2960
Case No.
Diagnosis
Bilateral renal agenesis Bilateral renal agenesis Bilateral renal agenesis Bilateral renal agenesis Bilateral renal agenesis Infantile polycystickidney Infantile polycystickidney Unilateral renal agenesis and contralateral nmlticystic kidney Unilateral renal agenesis and contralateral multicystic kidney Unilateral renal agenesis and contralateral muldcystic kidney Unilateral renal agenesis and contralateral multicystic kidney Unilateral renal agenesis and contralateral multicystic kidney Unilateral renal agenesis and contralateral multicystic kidney Preterm PROM (>5 wk) Preterm PROM (>8 wk) Thanatophoric dwarfism Thanatophoric dwarfism Thanatophoric dwarfism Thanatophoric dwarfism Thanatophoric dwarfism Thanatophoric dwarfism
Outcome
Stillbirth Neonatal Neonatal Stillbirth Neonatal Neonatal Neonatal Neonatal
death death death death death death
Stillbirth Neonatal death Neonatal death Neonatal death Neonatal death Stillbirth Neonatal death Neonatal death Neonatal death Neonatal death Neonatal death Neonatal death Neonatal death
PROM, Premature rupture of membranes.
with lethal pulmonary hypoplasia died in the first 24 hours after birth, and the diagnosis was confirmed at autopsy by lung weight/body weight ratios and radial alveolar counts. 9 The measurements of the groups with preterm premature rupture of membranes versus those with pulmonary hypoplasia were compared by Fisher's exact and two-tailed t tests. A value ofp < 0.01 was considered statistically significant. To determine the relationship between each parameter and lung weight or lung weight/body weight ratio, three patients (anhydramnios without pulmonary hypoplasia, hydrocephalus, and Turner syndrome) who could undergo measurements for the eight parameters were included, and regression analyses were performed. Results There was a significant linear relationship between thoracic circumference and gestational age (y -- 0.834x2.477 [R2= 0.948, p < 0.001]), between thoracic area and gestational age (y= 2.846x- 42.54 [R2=0.929, p < 0.001]), between thoracic area minus heart area and gesrational age (y= 2.128x- 31.29 JR2= 0.906, p < 0.001]), and between lung area and gestational age (y = 1.074x15.78 [R2= 0.866, p < 0.001]). Regression analyses of the relationships among the four ratios (thoracic circumference/abdominal circumference, thoracic area/heart area, thoracic area minus heart area/thoracic area, and lung area/thoracic area)
and gestational age revealed these ratios to be relatively constant. The mean value was 0.888 (SD 0.044) for the thoracic circumference/abdominal circumference ratio, 4.05 (SD 0.51) for the thoracic area/heart area ratio, 0.744 (SD 0.034) for the thoracic area minus heart area/thoracic area ratio, and 0.403 (SD 0.052) for the lung area/thoracic area ratio. After the establishment of nomograms, the efficacy of parameters was assessed by studying 30 patients with premature rupture of membranes (gestational ages 27 to 34 weeks) and 21 patients with pulmonary hypoplasia (gestational ages 20 to 39 weeks). The measurements for these patients were compared with the nomograms (-2 SD lines) established from the patients with intact membranes. Thoracic circumference was at or below the - 2 SD line in 5 of the 30 patients with premature rupture of membranes (16.7%) and in all 21 with pulmonary hypoplasia (100%) (p < 0.01). Thoracic area was at or below the - 2 SD line in 4 of the 30 with premature rupture of membranes (3.3%) but in all 21 with pulmonary hypOplasia (100%) (p < 0.01). Thoracic area minus heart area was at or below the - 2 SD line in 4 of the 30 with premature rupture of membranes (13.3%) and in all 6 with pulmonary hypoplasia (100%) (p < 0.01), and lung area was at or below the - 2 SD line in none of the 30 with premature rupture of membranes (0%) but in 13 of the 16 with pulmonary hypoplasia (81.3%) (p < 0.01) (Figs. 2 and 3). The thoracic circumference/abdominal circum-
480
Yoshimura et al.
August 1996
AmJ Obstet Gynecol
35"
9080-
30"
70-
~ - 2 S D
25-
0 o 8
oo°
20-
60" .
o. ...
20i0-
10
0 20
25
30
35
40
/
. 8" u 20
gestational age (weeks) 7060-
•
/ °o
20-
10- o._~-----"~•
°
-2SD
8 e •
.• .
•
e iw
i u 30 35 gestationat age (weeks)
i
40
/
30-
o o ~ o o
•
i 25
35-
= 4o~o
D
o'l
40-
u~ 50 <
30-
S
30"
• ..
15"
~
50" 40"
E 25< .n 2015-
•
I0-
•
.
50
20 '
2'5 30 ' 35 ' gestational age (weeks)
40 '
20
25 30 gestational age
35
(weeks)
40
Fig. 2. Distribution of thoracic circumference (TC), thoracic area (TA), thoracic area minus heart area (HA), and lung area (LA) in preterm premature rupture of membranes and pulmonary hypoplasia. Open circles,Preterm premature rupture of membranes; solid circles,pulmonary hypoplasia. Values <-2 SD were considered abnormal.
Table II. Mean _+ SD of gestational a g e - i n d e p e n d e n t parameters in three groups
Control Preterm PROM Pulmonary hypoplasia
TC/AC
TA/HA
TA- HA/TA
LA/TA
0.888 + 0.044* 0.887 - 0.054* 0.748 _ 0.076*
4.05 -+ 0.51 t 4.05 + 0.48]3.15 + 0.84]-
0.744 + 0.0345 0.742 _+0.026 + 0.664 +_0.076 +
0.403 -+ 0.052 0.420 -+ 0.079 0.351 + 0.096
TC, Thoracic circumference; AC, abdominal circumference; TA, thoracic area; HA, heart area; LA, lung area; PROM, premature rupture of membranes. *Control and preterm PROM versus pulmonary hypoplasia, p < 0.01. tControl and preterm PROM versus pulmonary hypoplasia, p < 0.01. $Control and preterm PROM versus pulmonary hypoplasia, p < 0.01.
ference ratio was 0.887_+ 0.054 in preterm premature rupture of m e m b r a n e s but 0.748-+ 0.076 (p< 0.01) in p u l m o n a r y hypoplasia, the thoracic area/heart area ratio was 4.05 + 0.048 in preterm premature rupture of m e m branes but 3.15 + 0.084 in p u l m o n a r y hypoplasia (p< 0.01), the thoracic area minus heart area/thoracic area ratio was 0.742 _+ 0.026 in preterm premature rupture of m e m b r a n e s but 0.664 + 0.079 in p u l m o n a r y hypoplasia ( p < 0 . 0 1 ) , and the lung area/thoracic area ratio was 0.420_+ 0.079 in preterm premature rupture of m e m branes but 0.351_+0.096 in p u l m o n a r y hypoplasia (Table II). The diagnostic index values of the eight ultrasonographic parameters for prediction of lethal p u l m o n a r y hypoplasia are shown in Table III. Lung area (gestational
a g e - d e p e n d e n t parameter) and thoracic circumfere n c e / a b d o m i n a l circumference ratio (gestational agei n d e p e n d e n t parameter) exhibited the highest diagnostic accuracy (sensitivity 81.3% and 90.5%, specificity 100% and 90.0%, positive predictive value 100% and 86.4%, and negative predictive value 90.9% and 93.1%, respectively). There was a significant linear relationship between lung weight and the gestational a g e - d e p e n d e n t parameters (thoracic circumference [y= 2 . 0 5 x - 20.50, / U = 0.613], thoracic area [y = 0 . 0 7 4 x - 2 . 9 8 , / ~ = 0.568], thoracic area m i n u s heart area [y = 0 . 1 x - 3.152, R 2 = 0.731], and lung area [y= 0 . 2 4 x - 6.021, / ~ = 0.786]) (Fig. 4). There was also a significant linear relationship between the lung w e i g h t / b o d y weight ratio and the gestational
Yoshimura et al.
Volume 175, N u m b e r 2 A m J O b s t e t Gynecol
1.1-
7-
1"
.£
88o80
~o
0.9"
o 0.8-
-w;----, "
00
6-
"+2SD
o
o 5-
°800
~
~
~
o o o ° O°•o~g o• O0
2i
3i0 35 gestati0nal age (weeks)
•
e
0.6-
2'5
~
24-
2sD
i
~
o
3-
20
/
•
0.7-
0.5
481
i
410
20
i
+2SD o o
o
i
*-
i
i
25 30 35 gestational age (weeks)
40
0.6"
+2SD 0.5-
.£ 0.9" .2
0.4" < E~ 0.3-
+2SD
0.8• o
@
o.7-
• •
°
•
•
• •
L.-----2SD
"---~-2sD
0.2"
0.60.5
oooo •
0.1-
%
i
25
i
30
~'s
0
i
40
gestationaI age {weeks)
2'o
~5
,'0
&
i
40
gestational age (weeks)
Fig. 3. Distribution of thoracic circumference (TC)/abdominal circumference (AC), thoracic area (TA)/heart area (HA), thoracic area minus heart area/thoracic area, and lung area (LA)/thoracic area ratio in preterm premature rupture of membranes and pulmonary hypoplasia. Open circles, Preterm premature rupture of membranes; solid circles, pulmonary hypoplasia. Values <-2 SD were considered abnormal. Table III. Diagnostic i n d e x values of lethal p u l m o n a r y hypoplasia
Gestational age dependent TC (n = 51) TA (n = 51) TA - HA (n = 46) LA (n = 46) Gestational age independent TC/AC (n = 51) TA/HA (n = 46) T A - HA/TA (n = 46) LA/TA (n = 46)
Sensitivity (%)
Specificity(%)
PPV (%)
NPV (%)
Accuracy (%)
100 100 100 81.3
83.3 86.7 86.7 100
80.8 84.0 80.0 100
100 100 100 90.9
90.2 92.2 91.3 93.5
90.5 68.8 68.8 31.3
90.0 100 96.7 100
86.4 100 91.7 100
" 93.1 85.7 85.3 73.2
90.2 89.1 87.0 76.1
PPV, Positive predictive value; NPV, negative predictive value; °o-. age-independent parameters (thoracic circumfere n c e / a b d o m i n a l c i r c u m f e r e n c e ratio [y = 0 . 0 7 8 x - 0.049, R 2 = 0.651], t h o r a c i c a r e a / h e a r t a r e a ratio [y = 0 . 0 0 4 9 x 0 . 0 0 6 2 , / ~ = 0.596], t h o r a c i c area m i n u s h e a r t a r e a / t h o racic a r e a ratio [y = 0 . 0 4 8 x - 0.022, R 2 = 0.40], a n d l u n g a r e a / t h o r a c i c area ratio [y=O.O295x-O.O012, R~= 0.186]) (Fig. 5). Comment
T h e p r e s e n c e of p u l m o n a r y hypoplasia has b e e n well d o c u m e n t e d in fetuses with l o n g - s t a n d i n g o l i g o h y d r a m nios, w h i c h c o u l d b e t h e r e s u l t o f c h r o n i c a m n i o t i c fluid leakage o r fetal r e n a l disease or it c o u l d b e idiopathic.
T h e e x t e n t o f fetal p u l m o n a r y hypoplasia d e p e n d s o n t h e degree and the duration of oligohydramnios and on the gestational age a n d stage o f l u n g d e v e l o p m e n t w h e n it occurs. F r o m t h e clinician's s t a n d p o i n t , it is desirable to b e able to distinguish hypoplasia as e i t h e r l e t h a l or n o n lethal. T h e c u r r e n t study was u n d e r t a k e n to clarify t h e clinical diagnosis o f p u l m o n a r y hypoplasia i n surviving infants, with t h e p r e d i c t i o n o f t h e l e t h a l f o r m o f p u l m o n a r y hypoplasia u s e d as a n e n d point. Previous investigators have a t t e m p t e d to achieve t h e a n t e n a t a l diagnosis of the lethal a n d n o n l e t h a l f o r m s o f p u l m o n a r y hypoplasia by use of n o m o g r a m s o f c h e s t c i r c u m f e r e n c e m e a s u r e m e n t s t h r o u g h o u t gestation 4 or
482
Yoshimura et al.
August 1996 AmJ Obstet Gynecol
40 n=10 r=0.89(p<0.001) y=2.41x-6.02
30
O
bJ0 O
20
ooo
10
0
•
•
4
6
I
I
I
I
I
8
10
12
14
16
LA(cm2) Fig. 4. Relationship between lung area (LA) and lung weight (LW).
0.030
0 n=17 0.025
0
iza
0.020 0.015
OO ~ ~ OO
0.010 0.005
O
/;9/°
0 0.65
°
O
I
I
I
I
I
I
0.70
0•75
0.80
0.85
0.90
0.95
T C / A C ratio Fig. 5. Relationship between thoracic circumference (TC)/abdominal circumference (AC) ratio and lung weight (LW)/body weight (BW) ratio. the chest c i r c u m f e r e n c e / a b d o m i n a l c i r c u m f e r e n c e ratio 5, 6 and the chest area minus heart a r e a / c h e s t area ratio, 7 which have the advantage of being constant t h r o u g h o u t gestation. T h e use of chest circumference, however, requires accurate knowledge of the length of gestation. T h e r e f o r e these n o m o g r a m s c a n n o t be used in fetuses with unknown gestational age or fetuses with suspected intrauterine growth restriction, which is frequently present• T h e chest c i r c u m f e r e n c e / a b d o m i n a l circumference ratio is gestational a g e - i n d e p e n d e n t and can be used reliably without accurate knowledge of the gestational age. T h e major limitation of this m e t h o d is that it c a n n o t be used in fetuses with large chest circumferences caused by the
excessive skin and subcutaneous tissue in disorders that affect or distort a b d o m i n a l c i r c u m f e r e n c e m e a s u r e m e n t s (e,g., fetuses with polycystic kidneys, obstructive uropathy, or omphalocele). It is known t h a t many cases of p u l m o n a r y hypoplasia are due to p r o l o n g e d r u p t u r e of membranes. Thus the use of the two most c o m m o n parameters (chest circumference and chest circumfere n c e / a b d o m i n a l circumference ratio) is questionable because the effect of p r e m a t u r e r u p t u r e of m e m b r a n e s on these m e a s u r e m e n t s has n o t b e e n investigated. With the above considerations in mind, we a t t e m p t e d in this cross-sectional study to establish n o m o g r a m s of ultrasonographic parameters that could possibly reflect fetal lung mass by studying 150 u n c o m p l i c a t e d pregnan-
Volume 175, Number 2 AmJ Obstet Gynecol
cies with intact membranes at the examination. The relationships between gestational age and thoracic circumference, thoracic area, thoracic area minus heart area, and lung area showed significant differences, but the four ratios (thoracic circumference/abdominal circumference, thoracic a r e a / h e a r t area, thoracic area minus heart area/thoracic area, and lung area/thoracic) were relatively constant. The chest circumference/abdominal circumference ratio was found to differ slightly from previously reported norms (mean value 0.91 to 0.94 in previous studies vs 0.88 in the current study). This difference can be explained by the fact that we measured the bony chest circumference (the method of Vintzileos et al. 7) instead of measuring from the outer edge to the outer edge of the fetal thorax as done previously. Although we used the method of Vintzilens et al., our results for the thoracic circumference/abdominal circumference ratio, the thoracic a r e a / h e a r t area ratio, and the thoracic area minus heart a r e a /t h o racic area ratio were somewhat higher than their norms. This difference may be due to the difference in body structure between Japanese and American fetuses. In patients with premature rupture of membranes the ruptured membrane itself is known to cause fetal compression and therefore may be inherently associated with alterations in fetal biometry. The current study addressed this point by evaluating fetuses with ruptured membranes who were unlikely to have pulmonary hypoplasia because the duration of premature membrane rupture was <1 week. In all cases with premature rupture of membranes the measurements were within the normal range for all three gestational a g e - i n d e p e n d e n t ratios. The fact that the - 2 SD line of the four ratios was not altered by the presence of premature rupture of membranes makes the ratios potentially valuable tools for evaluating fetuses with prolonged premature rupture of membranes or uncertain gestational age. As noted in Table III, most parameters had positive and negative predictive values that were too high to be clinically useful. The diagnostic accuracy of the studied parameters was much higher than that of previously reported parameters. 7 This discrepancy may be due to (1) differences in the populations studied, (2) inclusion of fetuses at particularly high risk for development of pulmonary hypoplasia in this study (all had documented oligohydramnios of >5 weeks' duration), (3) differences regarding the end point of prediction, which in our study was the lethal form of the disease, and (4) the difference in body structure between Japanese and American fetuses. In the current study the gestational development parameter lung area showed a significant linear relationship with lung weight, and the thoracic circumference/abdominal circumference ratio was significantly related to
Yoshimura et al. 483
the lung weight/body weight ratio. These relationships suggested that the thoracic circumference/abdominal circumference ratio was the best clinical parameter for the prediction of pulmonary hypoplasia. Recently several new techniques for the prediction of pulmonary hypoplasia have been reported, including fetal breathing movement, I° ductus arteriosus flow velocity modulation, 1~ and fetal upper respiratory tract function, I2 by Doppler ultrasonography. These methods have not been firmly established and thus require further investigation. In conclusion, lung area and the thoracic circumfere n c e / a b d o m i n a l circumference ratio had good diagnostic accuracy (93.5% and 90.3%) for antenatal diagnosis of pulmonary hypoplasia. This ratio shows significant linear relationships with lung weight (lung area) and the lung weight/body weight ratio (thoracic circumference/abdominal circumference ratio). These data suggested that the application of lung area and the thoracic circumfere n c e / a b d o m i n a l circumference ratio are clinically useful for the evaluation of fetal pulmonary hypoplasia. REFERENCES
1. Potter EL. Bilateral renal agenesis. J Pediatr 1946;29:68-73. 2. Thomas IT, Smith DW. Oligohydramnios, cause of the nonrenal features of Potter's syndrome, including pulmonary hypoplasia. J Pediatr 1974;84:811-4. 3. Nimrod C, Varela Gittings F, Machin G, Campbell D, Wesenberg R. The effect of very prolonged membrane rupture on fetal development. AmJ Obstet Gynecol 1984;148:540-4. 4. Nimrod C, Davies D, Iwanicki S, Harder J, Persaud D, Nicholson S. Ultrasound prediction of pulmonary hypoplasia. Obstet Gynecol 1986;68:496-8. 5. Fong K, Ohlsson A, Zalev A. Fetal thoracic circumference: a prospective cross-sectional study with real-time ultrasound. AmJ Obstet Gynecol 1988;158:1154-60. 6. Johnson A, Callan N, Bhutani VK, Colmorgen GHC, Weiner S, Bolognese RJ. Ultrasonic ratio of fetal thoracic to abdominal circumference: an association with fetal pulmonary hypoplasia. AmJ Obstet Gynecol 1987;157:764-9. 7. Vintzileos AM, Campbell WA, Rodis JF, Nochimson DJ, Pinette MG, Petrikovsky BM. Comparison of six different ultrasonographic methods for predicting lethal fetal pulmonary hypoplasia. AmJ Obstet Gynecol 1989;161:606-12. 8. Yoshimura S, Masuzaki H, Fukuda H, Yamabe T. Comparison of eight different ultrasonographic parameters for predicting lethal fetal pulmonary hypoplasia. Jpn J Med Ultrasonics 1992;19:207-13. 9. Askenazi SS, Perlmma M. Pulmonary hypoplasia: lung weight and radial count as criteria of diagnosis. Arch Dis Child 1979;54:614-8. 10. Blott M, Greenough A, Nicolaides KH, Moscoso G, Gibb D, Campbell S. Fetal breathing movements as predictor "of favourable pregnancy outcome after oligohydramnios due to membrane rupture in second trimester. Lancet 1987;1: 129-31. 11. EyckJV, Mooren K, WladimiroffJW. Ductus arteriosus flow velocity modulation by fetal breathing movements as a measure of fetal lung development. Am J Obstet Gynecol 1990; 163:558-66. 12. Fox HE, Badalian SS, Timor-Tritsch TE, Stolar CJH. Fetal upper respiratory tract function in cases of antenatally diagnosed congenital diaphragmatic hernia: preliminary observations. Ultrasound Obstet Gynecol 1993;3:164-7.