OBJECTIVE: Neonatal respiratory distress syndrome (RDS) affects approximately 1% of live births, and the probability of RDS continues to be a major determinant in the timing of delivery. This study was designed to investigate the optimal gestational age–specific cutoff value for a surfactant-to-albumin ratio assay for predicting RDS.
METHODS: Amniotic fluid surfactant-to-albumin ratio data were collected prospectively for a 2-year period. Women were included in the study if they delivered within 72 hours of surfactant-to-albumin ratio estimation. RDS was defined by the presence of 2 or more of the following criteria: evidence of respiratory compromise shortly after delivery and a persistent oxygen requirement for more than 24 hours, administration of exogenous pulmonary surfactant, and/or radiographic evidence of hyaline membrane disease.
RESULTS: A total of 415 mother–neonate pairs (28 RDS, 387 non-RDS) met criteria for analysis. Both gestational age and surfactant-to-albumin ratio values were independent predictors of RDS. By modeling the odds of RDS by using a logistic regression with gestational age and surfactant-to-albumin ratio values as continuous variables, a probability of RDS of 15% or less can be achieved with a surfactant-to-albumin ratio cutoff of 60 mg or more surfactant/g albumin at 28 weeks of gestation, 50 or more at 30 weeks, 40 or more at 33 weeks, 30 or more at 35 weeks, and 20 or more at 37 weeks.
CONCLUSIONS: These data describe a means of stratifying the probability of neonatal RDS using both gestational age and surfactant-to-albumin ratio value and may be a useful model for clinical decision-making.
LEVEL OF EVIDENCE: II-2
Gestational age and surfactant-to-albumin ratio assay are independent predictors of neonatal respiratory distress syndrome and should be used in combination in clinical decision making.
From the Division of Maternal–Fetal Medicine, Department of Obstetrics, Gynecology and Reproductive Biology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts.
Received August 14, 2003. Received in revised form October 26, 2003. Accepted December 3, 2003.
Address reprint requests to: Dr. Thomas F. McElrath, Harvard Medical School, c/o Division of Maternal–Fetal Medicine, Department of Obstetrics, Gynecology & Reproductive Biology, Brigham & Women's Hospital, 75 Francis Street, Boston, MA 02115; e-mail: email@example.com.
Neonatal respiratory distress syndrome (RDS) refers to respiratory compromise presenting at or shortly after birth specifically as a result of a deficiency of pulmonary surfactant, an endogenous detergent that serves to decrease the surface tension within alveoli, thereby preventing alveolar collapse. Neonatal RDS affects approximately 1% of all live births1–3; however, not all infants are at equal risk. The pulmonary system is among the last of the fetal organ systems to become functionally mature. As such, RDS is primarily—although not exclusively—a disease of premature infants with an incidence and severity that is highly dependent on gestational age.1–4
To assist obstetric care providers in their counseling of pregnant women at risk of preterm delivery, a series of biochemical tests has been developed in an attempt to modify an individual fetus's gestational age–related risk of developing RDS. All such tests rely on the direct or indirect measurement of surfactant phospholipids secreted by the fetal lung into amniotic fluid.2,5–7 One of the most widely used tests for fetal lung maturity is the TDx-FLM II surfactant-to-albumin assay (Abbott Laboratories, Abbott Park, IL), which uses fluorescence polarization to determine the relative concentrations of surfactant and albumin in amniotic fluid; results are given as mg of surfactant per 1 g of albumin.
Despite the close relationship between RDS and gestational age,1–4 biochemical tests for fetal lung maturity have traditionally been interpreted as being dichotomous (ie, either “positive” or “negative”). This approach is fundamentally misleading. The rapid decline in the incidence of RDS within increasing gestational age—from 30–50% for fetuses delivered before 32 weeks of gestation to approximately 1% at term1—results in poor performance for all fetal lung maturity tests at higher gestational ages.8 What is needed to improve obstetric care is not an innovation in biochemical assay technique but rather an improved understanding of the probability of a fetus developing RDS for a given test value at a specific gestational age.8,9
In 1994, Tanasijevic et al10 published the first quantitative modification of a fetal lung maturity test to adjust the gestational age-related risk of a fetus developing RDS. In this publication, the authors developed a logistic equation to estimate the probability of neonatal RDS as a function of gestational age and the TDx-FLM II surfactant-to-albumin ratio Advantages of the surfactant-to-albumin ratio assay are that it is a rapid test that is relatively easy to perform and highly reproducible. A second-generation assay (TDx-FLM II) has been developed with improved sample handling characteristics, but this test has not been systematically examined in relation to gestational age. This study was therefore designed to derive predictive equations that will allow the risk of neonatal RDS to be defined both as a function of the TDx-FLM II surfactant-to-albumin ratio assay and gestational age.
MATERIALS AND METHODS
All singleton, nonanomalous, liveborn infants born to women who underwent fetal lung maturity screening at the Brigham & Women's Hospital in Boston, Massachusetts, between January 1998 and January 2000 with the TDx-FLM II surfactant-to-albumin ratio assay marketed by Abbott Laboratories were identified. The hospital's institutional review board approved this study. Patients were included only if TDx-FLM II surfactant-to-albumin ratio measurements were obtained within 72 hours of delivery. Maternal records were matched to neonatal charts and abstracted. Maternal charts were reviewed for gestational dating criteria, timing, and results of TDx-FLM II surfactant-to-albumin ratio testing, date of delivery, smoking and diabetic status, maternal race, and antenatal corticosteroid status. Neonatal records provided information on infant sex, results of chest radiograph, and oxygen requirements if present. Multiple gestations were excluded from this analysis.
Gestational age was determined by in vitro fertilization or intrauterine insemination dating, a first-trimester ultrasound, or last menstrual period confirmed by a second-trimester ultrasound, as available. Cases dated only by a second- or third-trimester ultrasound were excluded. Maternal race was self-described in the medical record. Maternal diabetes was defined as any insulin requirement, including insulin-dependent pregestational diabetes and insulin-requiring gestational diabetes. Previous publications have shown that the TDx-FLM II surfactant-to-albumin ratio assay performs equally well in diabetic and nondiabetic pregnancies in terms of predicting risk of neonatal RDS11,12; as such, data from diabetic and nondiabetic gestations were analyzed together. Preeclampsia was defined as the presence of new-onset persistent hypertension (blood pressure 140 mm Hg or more systolic and/or 90 mm Hg diastolic) and new-onset proteinuria (300 mg or more urinary protein per 24 hours) after 20 weeks of gestation. Chronic hypertension was defined by maternal antihypertensive use before pregnancy. Maternal smoking was identified in the medical record.
Neonatal RDS was diagnosed by the presence of at least 2 of the following 3 criteria: 1) evidence of respiratory compromise (tachypnea, retractions, and/or nasal flaring) shortly after delivery and a persistent oxygen requirement for more than 24 hours, 2) administration of exogenous pulmonary surfactant, and/or 3) radiographic evidence of neonatal pulmonary hyaline membrane disease as diagnosed by an attending pediatric radiologist or neonatologist. Radiographic evidence of neonatal RDS includes atelectasis, air bronchograms, and a diffuse reticulogranular infiltrate.2 Cases of transient tachypnea of the newborn and suspected neonatal pneumonia were excluded.
Amniotic fluid specimens were analyzed with the Abbott TDx-FLM II surfactant-to-albumin ratio assay as previously described.13 Briefly, the TDx-FLM II surfactant-to-albumin ratio assay predicts surfactant levels based on the deflection of polarized light within an aliquot of amniotic fluid; “TDx-FLM II” represents a brand name of Abbott Laboratories. Because the results of fetal lung maturity screening are not significantly affected by the source of amniotic fluid,14 vaginal pool and amniocentesis specimens were analyzed jointly.
Statistical analysis consisted of χ2 analysis, Spearman's correlation coefficient, and multivariable logistic regression with the STATA statistical package (STATA Corporation, College Station, TX). Using logistic regression, a prediction rule for the probability of neonatal RDS was developed with both the TDx-FLM II surfactant-to-albumin ratio value and gestational age as the descriptive variables.
A total of 415 mother–neonate pairs met criteria for analysis and were abstracted. As evidenced in Figure 1, the gestational ages at delivery among eligible pregnancies were closely clustered around the median of 37.0 weeks (25th percentile: 35.3 weeks and 75th percentile: 38.1 weeks). Twenty-eight cases of neonatal RDS were confirmed, for an incidence of 6.7% (95% confidence interval [CI] 4.5%, 9.6%). As expected, the incidence of RDS was closely related to gestational age. The incidence of neonatal RDS was 26.3% (95% CI 9.1%, 51.2%) for deliveries 32 weeks or less of gestation, 15% (95% CI 7.4%, 25.7%) between 33 and 34 weeks, 6.9% (95% CI 3.0%, 13.1%) between 35 and 36 weeks, and 2.3% (95% CI 0.07%, 5.4%) for deliveries at or after 37 weeks.
Figure 2 displays the distribution of pregnancies diagnosed with and without neonatal RDS as a function of gestational age and TDx-FLM II surfactant-to-albumin ratio value. The degree of scatter is marked, but a tendency for TDx-FLM II surfactant-to-albumin ratio value to increase with increasing gestational age is evident (Spearman's r = 0.31). All but 1 case of neonatal RDS (in an infant delivered to an A2 diabetic mother at 37.4 weeks of gestation who required 5 days of oxygen hood but no surfactant or intubation) occurred in a pregnancy with a TDx-FLM II surfactant-to-albumin ratio value of less than 70 mg/g, the cutoff value recommended by the manufacturers and endorsed by the American College of Obstetricians and Gynecologists2 that is predictive of the absence of RDS irrespective of gestational age with a sensitivity of at least 85%.
Table 1 presents the maternal clinical characteristics of the study population and their association with neonatal RDS. Within the limits of our sample size, no significant differences were evident within these subcategories (Table 1). Administration of antenatal corticosteroids to pregnancies at risk has been shown to decrease the incidence of neonatal RDS (as well as intraventricular hemorrhage and necrotizing enterocolitis) by approximately 50%.15 Because antenatal corticosteroid therapy is only recommended for pregnancies threatening to deliver before 34 weeks of gestation16,17 and the median gestational age at delivery (and hence of testing) in the study population was in the 37th week, it is likely that antenatal corticosteroids in this study is a proxy for early gestational age. In a bivariate analysis of the 23.9% (99 of 415) pregnancies in the study population exposed to antenatal corticosteroids, 14.1% (14 of 99) of those exposed to antenatal corticosteroids were diagnosed with neonatal RDS as compared with 4.4% (14/316) of those not exposed (χ2 P = .001). However, when corticosteroid exposure was modeled in a logistic regression controlling for gestational age and TDx-FLM II surfactant-to-albumin ratio value, the odds ratio (OR) indicated a nonsignificant effect (OR 0.53; 95% CI 0.18, 1.51).
To better investigate the relationship between gestational age and TDx-FLM II surfactant-to-albumin ratio values, the odds of neonatal RDS were modeled by using a logistic regression with gestational age and TDx-FLM II surfactant-to-albumin ratio values entered as continuous variables. The β coefficient was 9.30 for the constant, .26 for gestational age, and .06 for the TDx-FLM II surfactant-to-albumin ratio. The ORs derived from the model are displayed in Table 2. Additionally, Table 2 presents the ORs for African-American race, maternal smoking, antenatal steroid administration, preeclampsia, and insulin-requiring diabetes with respect to the risk of neonatal respiratory distress. None of these additional risk factors significantly predicted the risk of RDS when included in separate models that controlled for gestational age and TDx-FLM II surfactant-to-albumin ratio. We therefore used the significant gestational age and TDX-FLM II surfactant-to-albumin ratio coefficients to derive a predictive equation yielding a percent risk of neonatal RDS specific for each week of gestational age and TDx-FLM II surfactant-to-albumin ratio value as displayed in Table 3.
In this analysis, we offer a gestational age- and test-specific probabilistic method of determining the risk of neonatal RDS during pregnancy. Timing of delivery remains a major area of debate in the discipline of obstetrics. In counseling women about the timing of delivery, obstetric care providers must balance the risk of prematurity (including neonatal RDS) against the potential maternal and fetal risks associated with continuing the pregnancy. To date, screening for fetal lung maturity with tests such as the TDx-FLM II surfactant-to-albumin ratio assay have proven a valuable aid to obstetric care providers. However, the utility of this information is limited by the current practice of interpreting the result as a simple bivariate positive (“mature,” ie, predictive of the absence of RDS, or negative, ie, “immature”). In any disease with a developmental stage–specific incidence that changes rapidly over a short period of time during pregnancy,1 a screening test will yield both high false-positive and false-negative rates depending on the gestational age at which the investigation is performed.
Using decision-analytic techniques, several authors18,19 have attempted to guide the application of fetal lung maturity testing by recommending that such testing be performed only when the results would be most likely to accurately guide obstetric recommendations. Because the performance characteristics of these tests are so closely related to gestational age, these investigators have suggested using such tests only during that gestational age window where the performance characteristics are more uniform (ie, between 34 and 36 weeks of gestation). Our findings offer an alternative and clinically contemporaneous method for practitioners to evaluate the risk of neonatal RDS across a wide range of gestational ages given the underlying physiology of fetal lung development with advancing gestational age. A TDx-FLM II surfactant-to-albumin ratio value of 70 mg/g or more was originally recommended by the manufacturers as a cutoff value denoting a reduced risk of fetal lung immaturity because—in a crude analysis that did not take into account gestational age—it represented a 15% risk of neonatal RDS among all test subjects. It has since become clear that a single cutoff value for the TDx-FLM II assay3,10 (or other screening test for fetal lung maturity20) for all gestational ages is not appropriate. In our model, where we stratified risk of RDS by gestational age and TDx-FLM II surfactant-to-albumin ratio value, this same level of risk was evident for a TDx-FLM II surfactant-to-albumin ratio value of approximately 20 mg/g at 37 weeks, 30 mg/g at 35 weeks, 40 mg/g at 33 weeks, 50 mg/g at 30 weeks, and 60 mg/g at 28 weeks (Table 3).
In addition to gestational age, several maternal–fetal characteristics may influence the risk of a given fetus developing RDS including, among other factors, such factors as maternal race,21 preeclampsia,22 and intrauterine exposure to cigarette smoke23 and cocaine.24 In this study, we included several characteristics in the analysis (including maternal race, smoking, diabetes, preeclampsia, and corticosteroid administration) and did not find a significant statistical contribution by these characteristics. Although the weight of evidence in the literature suggests that these characteristics do indeed modify fetal lung maturity and, as such, affect the probability of RDS, we submit that the magnitude of these effects is relatively small compared with the overwhelming effect of gestational age and surfactant production as measured by the TDx-FLM II surfactant-to-albumin ratio assay. As such, these effects will factor less significantly into the overall prediction of lung maturity. The possible exception to this statement is antenatal corticosteroid administration. Unfortunately, late third-trimester pregnancies were disproportionately represented in our study population (median gestational age at delivery, 37 weeks; Figure 1), whereas it is pregnancies remote from term (less than 34 weeks) that have been shown consistently to benefit from antenatal corticosteroid therapy.16,17 As such, antenatal corticosteroid administration did not appear to be a statistically meaningful factor in predicting the likelihood of a fetus developing RDS. With additional sample size, further analysis of the role of antenatal corticosteroid administration on the probability of neonatal RDS at a given gestational age will be possible.
In summary, these data offer a means to stratify and individualize the risk of neonatal RDS by gestational age and TDx-FLM II surfactant-to-albumin ratio value. For example, a probability of RDS of 15% or less can be achieved with a TDx-FLM II surfactant-to-albumin ratio cutoff value of 60 mg or more surfactant/g albumin at 28 weeks of gestation, 50 or more at 30 weeks, 40 or more at 33 weeks, 30 or more at 35 weeks, and 20 or more at 37 weeks. We expect that these data will improve the understanding among the obstetric community that the probability of neonatal RDS is not a simple binary decision (“mature” or “immature”) but rather a gradation of risk dependent on the probabilistic nature of the screening test given the stage of fetal maturation. For some conditions, such as severe preeclampsia or clinical chorioamnionitis, the risks to the mother and fetus of expectant management clearly outweigh the potential benefits. As such, delivery should be considered regardless of gestational age or TDx-FLM II surfactant-to-albumin ratio value. In other less urgent conditions, however, an accurate assessment of the probability of neonatal RDS will likely assist in clinical decision making.
1. Robertson PA, Sniderman SH, Laros RK Jr, Cowan R, Heilbron D, Goldenberg RL, et al. Neonatal morbidity according to gestational age and birth weight from five tertiary care centers in the United States, 1983 through 1986. Am J Obstet Gynecol 1992;166:1629–41; discussion 1641–5.
2. American College of Obstetricians and Gynecologists. Assessment of fetal lung maturity. ACOG Educational Bulletin 230. Washington, DC: ACOG; 1996.
3. Fantz CR, Powell C, Karon B, Parvin CA, Hankins K, Dayal M, et al. Assessment of the diagnostic accuracy of the TDx-FLM II to predict fetal lung maturity. Clin Chem 2002;761–5.
4. Copper RL, Goldenberg RL, Creasy RK, DuBard MB, Davis RO, Entman SS, et al. A multicenter study of preterm birth and gestational age-specific neonatal mortality. Am J Obstet Gynecol
5. Gluck L, Kulovich MV, Borer RC, Keidel WN. The interpretation and significance of the lecithin/sphingomyelin ratio in amniotic fluid [review]. Am J Obstet Gynecol
6. Kulovich MV, Hallman MB, Gluck L. The lung profile. I. Normal pregnancy. Am J Obstet Gynecol
7. American College of Obstetricians and Gynecologists. Fetal maturity assessment prior to elective repeat cesarean delivery. ACOG Technical Bulletin 98. Washington, DC: ACOG; 1991.
8. Richardson DK, Heffner LJ. Fetal-lung maturity: tests mature, interpretation not. Lancet
9. Pinette MG, Blackstone J, Wax JR, Cartin A. Fetal lung maturity indices: a plea for gestational age–specific interpretation—a case report and discussion. Am J Obstet Gynecol
10. Tanasijevic MJ, Wybenga DR, Richardson D, Greene MF, Lopez R, Winkelman JW. A predictive model for fetal lung maturity employing gestational age and test results. Am J Clin Pathol
11. Delgado JC, Greene MF, Winkelman JW, Tanasijevic MJ. Comparison of disaturated phosphatidylcholine and fetal lung maturity surfactant/albumin ratio in diabetic and nondiabetic pregnancies. Clin Chem
12. Tanasijevic MJ, Winkelman JW, Wybenga DR, Richardson DK, Greene MF. Prediction of fetal lung maturity in infants of diabetic mothers using FLM S/A and disaturated phosphatidylcholine tests. Am J Clin Pathol
13. Russell JC. A calibrated florescence polarization assay for assessment of fetal lung maturity [published erratum in Clin Chem 1988 34 207). Clin Chem
14. Lauria MR, Dombrowski MP, Delaney-Black V, Bottoms SF. Lung maturity tests: relation of source, clarity, gestational age and neonatal outcome. J Reprod Med
15. Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treatment for prevention of the respiratory distress syndrome in premature infants. Pediatrics
16. National Institutes of Health Consensus Statement. Effect of antenatal steroids for fetal maturation on perinatal outcomes. 1994;12:1–24.
17. American College of Obstetricians and Gynecologists. Antenatal corticosteroid therapy for fetal maturation. ACOG Technical Bulletin 210. Washington, DC: ACOG; 1998.
18. Myers ER, Alvarez JG, Richardson DK, Ludmir J. Cost-effectiveness of fetal lung maturity testing in preterm labor. Obstet Gynecol
19. Macones GA, Bader TJ, Asch DA. Optimizing maternal-fetal outcomes in preterm labour: a decision analysis. Br J Obstet Gynaecol
20. Creasy GW, Simon NV. Sensitivity and specificity of the L/S ratio in relation to gestational age [review]. Am J Perinatol
21. Berman S, Tanasijevic MJ, Alvarez JG, Ludmir J, Lieberman E, Richardson DK. Racial differences in the predictive value of the TDx fetal lung maturity assay. Am J Obstet Gynecol
22. Winn HN, Klosterman A, Amon E, Shumway JB, Artal R. Does preeclampsia influence fetal lung maturity? J Perinat Med
23. Lieberman E, Torday J, Barbieri R, Cohen A, Van Vunakis H, Weiss ST. Association of intrauterine cigarette smoke exposure with indices of fetal lung maturation. Obstet Gynecol
24. Hanlon-Lundberg KM, Williams M, Rhim T, Covert RF, Mittendorf R, Holt JA. Accelerated fetal lung maturity profiles and maternal cocaine exposure. Obstet Gynecol
© 2004 The American College of Obstetricians and Gynecologists
This article has been cited