- Email: [email protected]
Two-year infant neurodevelopmental outcome after single or multiple antenatal courses of corticosteroids to prevent complications of prematurity
American Journal of Obstetrics and Gynecology (2004) 191, 217e24
www.elsevier.com/locate/ajog
Two-year infant neurodevelopmental outcome after single or multiple antenatal courses of corticosteroids to prevent complications of prematurity Arsenio Spinillo,a,* Franco Viazzo,a Rossella Colleoni,a Alberto Chiara,b Rosa Maria Cerbo,b Elisa Fazzic Departments of Obstetrics and Gynecology,a Neonatal Intensive Care,b and Child Neuropsychiatry,c University of Pavia, Istituto di Ricovero e Cura a Carattere Scientifico Policlinico S. Matteo ed Istituto di Ricovero e Cura a Carattere Scientifico C. Mondino, Pavia, Italy Received for publication October 14, 2003; revised December 10, 2003; accepted December 22, 2003
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– KEY WORDS Betamethasone Dexamethasone Periventricular leukomalacia Prematurity
Objective: This study was undertaken to evaluate the effect of exposure to multiple antenatal steroid courses on short-term neonatal morbidity and 2-year infant neurodevelopmental outcome. Study design: This was a prospective observational study of 201 preterm singleton infants who received 1 or more courses of corticosteroids to prevent complications of prematurity and were delivered between 24 and 34 weeks’ gestation at a single institution. Neurodevelopmental outcome of the infants was evaluated at 2 years corrected age. Logistic regression analysis was used to perform multivariate analyses of associations and trends. Results: One hundred thirty-eight subjects (68.7%) received at least 1 complete course of betamethasone, whereas 63 (31.3%) patients were treated with dexamethasone. The prevalence of multiple steroid doses exposure was 26.8% (37/138) in betamethasone and 52.4% (33/63) in dexamethasone group. The prevalence of infant leukomalacia, including both prolonged echogenicity and cystic leukomalacia, was 25.9% (34/131) after a complete corticosteroid course, 40% (6/15) after 1, 42.3% (12/28) after 2, and 44.4% (12/27) after more than 2 additional courses, respectively (adjusted P for trend = .011). In the same categories of steroid exposure, the corresponding prevalences of 2-year infant neurodevelopmental abnormalities were 18% (20/111), 21.4% (3/14), 29.2% (7/24), and 34.8% (8/23), respectively (adjusted P for trend = .038). Multivariate study of first grade interaction suggested that the risk of leukomalacia and 2-year infant neurodevelopmental abnormalities associated with multiple doses exposure was confined to dexamethasone. In fact, compared with betamethasone, exposure to multiple doses of dexamethasone was associated with an increased risk of leukomalacia (19/33 compared with 11/37; odds ratio [OR] = 3.21, 95% CI = 1.07-9.77) and overall 2-year infant neurodevelopmental abnormalities (12/28 compared with 6/35; OR = 3.63, 95% CI = 1.03-13.58).
* Reprint requests: Arsenio Spinillo, MD, Department of Obstetrics and Gynecology, University of Pavia. Policlinico S. Matteo, P. le Golgi,19, 27100 Pavia, Italy E-mail: [email protected] 0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ajog.2003.12.023
218
Spinillo et al Conclusion: In this study, multiple antenatal courses of dexamethasone but not betamethasone were associated with an increased risk of leukomalacia and 2-year infant neurodevelopmental abnormalities. Ó 2004 Elsevier Inc. All rights reserved.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– A large number of controlled, observational studies and metanalyses support the use of antenatal corticosteroids to prevent neonatal complications of prematurity in women at risk of preterm delivery.1-3 A consensus panel of the National Institutes of Health in 1994 concluded that antenatal betamethasone or dexamethasone reduces the risk of neonatal death, respiratory distress syndrome, and intraventricular hemorrhage when administered up to 7 days before delivery.4 The same panel concluded that data were insufficient to establish a clear benefit beyond 7 days. A subsequent reconvening of the Consensus Conference cautioned against the use of repeated antenatal corticosteroids outside a randomized control trial.5 In recent years, it has been a common practice among obstetricians to prescribe weekly antenatal steroid courses to women undelivered after the first dose.6 In a recent randomized study, weekly courses of antenatal corticosteroids did not reduce short-term composite neonatal morbidity (perinatal death, respiratory distress syndrome, intraventricular hemorrhage, and periventricular leukomalacia) compared with a single course of treatment.7 In addition, animal data suggest that repeated antenatal steroid doses could interfere with the growth and development of the immature brain.8 Data on the effect of weekly antenatal corticosteroid courses on infant neurodevelopmental outcome are sparse and mainly related to antenatal betamethasone rather than dexamethasone administration.9,10 The purpose of this study was to evaluate the 2-year infant neurodevelopmental outcome in a cohort of women treated antenatally with single or repeated courses of either betamethasone or dexamethasone.
Methods This is a prospective observational study of all pregnant women delivered of a singleton live fetus between 24 and 34 weeks of gestation at a third level regional perinatal center between January 1, 1997, and December 3, 2000, who had received 1 or more antenatal courses of corticosteroid for the prevention of the complications of prematurity. Of a total of 315 singleton pregnancies delivered in the period examined; 16 severely malformed fetuses, 10 pregnancies with uncertain gestational age, 63 pregnancies delivered without corticosteroids treatment, and 25 with incomplete steroid administration (single dose) were excluded from the study. The remaining 201 women who received at least 1 complete antena-
tal steroid course constituted the study group. In these patients antenatal steroids were routinely administered either in the form of two 12-mg intramuscular doses of betamethasone, 24 hours apart, or dexamethasone, 6-mg intramuscular every 12 hours for 4 doses depending on the attending obstetrician. Subsequently, if the patient had not delivered, weekly repetitions of betamethasone or dexamethasone treatments were carried out at the discretion of the attending physician. Maternal data and clinical variables were collected prospectively and stored in a computer database. Preeclampsia was defined according to standard criteria.11 Fetal growth restriction was diagnosed when abdominal circumference ultrasonographic measurements decreased to below the 10th percentile of a standard growth curve for our population; the abnormal measurement of abdominal circumference was confirmed on at least 2 ultrasonographic examinations, 2 to 4 weeks apart, after the standard 18 to 22 weeks’ sonogram. To evaluate the severity of growth restriction, we also computed the birth weight ratio of each infant at any given gestational age. The birth weight ratio is the ratio between the observed and the expected birth weight, where expected birth weight is the 50th percentile of birth weight of our population corrected for the sex of the infant. The diagnosis of preterm premature rupture of membranes was based on clinical assessment, measurement of vaginal pH and ultrasonographic findings. Ritodrine was used to achieve tocolysis during the study period. The clinical diagnosis of chorioamnionitis was based on maternal temperature greater than 37.8(C plus 2 of the followings: (1) maternal tachycardia (O100 beats/min); (2) baseline fetal hearth rate greater than 160 beats/min; (3) positive endocervical or high-vaginal swab culture; (4) maternal white blood cell count greater than 18.000/ mm3; and (5) foul-smelling amniotic fluid. These criteria are similar to those recommended by Gibbs and Duff.12 After delivery all newborn infants were admitted to the neonatal intensive care unit of our department. A standard echoencephalogram was obtained within 24 hours of admission from all infants. Subsequently, serial cranial ultrasound was carried out every 48 to 72 hours during the first 2 weeks of life and weekly thereafter until they were discharged. Cranial ultrasounds were performed by 2 neonatologists (A.C. and R.M.C.) unaware of the type and dosages of antenatal steroid used. Infant intracranial hemorrhage was graded into 4 categories according to Papile et al.13 Periventricular leukomalacia included both infants with frank cystic de-
Spinillo et al generation of periventricular white matter and infants with persistently (more than a week) increased echodensity in the periventricular regions but not undergoing cystic degeneration. Neurodevelopmental evaluation of the infants was carried out by a child neuropsychiatrist not involved in the intensive care of the infants and unaware of maternal history. Examinations were carried out at discharge from hospital and at 3, 6, 12, and 24 months of corrected age. Neurologic evaluation of the newborn infants was based on the methods of AmielTison and Grenier.14 The Bayley scales of infant development were used to assess cognitive development (Mental Developmental Index [MDI]) at 12 to 24 months. Infants were grouped into 4 categories according to their final examination: (1) unimpaired = normal neurologic status and Bayley MDI O84; (2) minor impairment = abnormalities of tone or reflexes but functionally normal or borderline (71-84) Bayley MDI; (3) moderate impairment = spastic diplegia or hemiplegia with a Bayley MDI R71; and (4) severe impairment = spastic tetraplegia and/or severe mental retardation (Bayley MDI !71). This classification is similar to that proposed by De Vries et al.15 Among the 190 surviving infants treated antenatally with corticosteroids, 172 (90.5%) were followed-up throughout the study program. The 18 infants lost to follow-up had been judged neurologically normal at discharge from hospital. Of the 56 surviving infants who did not receive antenatal corticosteroids treatment, 51 (91.1%) were followed-up at 2 years of corrected age. Statistical analysis was carried out by using the Mann-Whitney U test or the c2 test to compare continuous and categorical variables, respectively. The c2 for trend was used to evaluate linear trends across ordered categories. Continuous variables are expressed as median and interquartile range when skewed or as mean and SD when normally distributed. Unconditional logistic regression was used to test for association between variables while adjusting for potential confounders and to test for linear trends and first-degree interaction between variables.16 Because this was not a randomized study, we did not compute a preliminary sample size. However, assuming a 20% prevalence of any given outcome (eg, overall intraventricular hemorrhage or leukomalacia) we had a 83% power to detect an odds ratio (OR) of 2 or more at the usual a values of .05.
Results The main antenatal characteristics of the studied population are reported in Table I. One hundred thirty-eight subjects (68.7%) received at least 1 complete course of betamethasone, whereas 63 (31.3%) patients were treated with dexamethasone. The median gestational age and birth weight were 31 weeks (interquartile
219 range = 4) and 1490 g (interquartile range = 670) in the betamethasone group and 31 weeks (interquartile range = 4) and 1456 g (interquartile range = 797) in the dexamethasone group (P = .88 and P = .25 by Mann-Whitney test compared with betamethasone, respectively). The median gestational age and birth weight of the 63 infants delivered without any antenatal steroid treatment mostly because of an emergency delivery and not presented in Table I were 33 weeks (interquartile range = 3) and 1759 g (interquartile range = 780) (P!.01 compared with treated infants for both comparisons). The mean values of birth weight ratio were 0.892 G 0.19 in the group of infants treated antenatally with betamethasone and 0.907 G 0.26 in those treated with dexamethasone (P = .94 by Mann-Whitney test). The corresponding value of birth weight ratio among untreated infants was 0.91 G 0.21. Patients who received dexamethasone were more likely to receive long-term tocolysis with ritodrine (c2 for trend = 4.14, P = .042) and multiple doses (c2 for trend = 15.7, P!.01). In addition, the median time interval from last steroid administration and birth was 3.2 days (range 1 to 7) in mothers receiving multiple and 5.7 days (range 1 to 28) in those receiving single doses (P!.001 by Mann-Whitney test). In Table II are reported the multivariate ORs of neonatal outcome variables adjusted by gestational age, birth weight ratio, ritodrine use, surfactant use, and multiple steroids dosages among infants born to mothers treated with dexamethasone as opposed to betamethasone. The occurrence of neonatal acidosis (arterial pH !7.2) both at birth and in the first 24 hours of life was significantly higher among infants delivered after antenatal dexamethasone administration compared with betamethasone treatment. Transient periventricular echogenicity, disappearing in the first week of life was recorded in 4 infants (6.3%) treated antenatally with dexamethasone and in 11 (8%) with betamethasone, respectively. After adjustment for potential confounders, the prevalence of periventricular leukomalacia was significantly higher among infants treated antenatally with dexamethasone as opposed to betamethasone (OR = 2.25, 95% CI = 1.07-4.74). The relative rates of tetraplegia and emiplegia-diplegia were 3.5 % (5/141) and 2.8% (4/141) in the betamethasone group and 3.6% (2/55) and 7.3% (4/55) in the dexamethasone group, respectively. In the group of 63 infants delivered without any steroid treatment, the prevalences of respiratory distress syndrome and neonatal death were 41.3% (26 of 63; OR = 1.58, 95% CI = 0.84-2.93 compared with treated infants) and 11.1% (7 of 63; OR = 2.16, 95% CI = 0.726.38 compared with treated infants). In addition, the prevalences of moderate-severe impairment and minor neurodevelopmental abnormalities were 7.8% (4/51) and 31.4% (16/51) (c2, 2 df = 7.12, P = .028 in comparison to treated infants), respectively. In logistic regression models, after adjustment for gestational age,
220
Spinillo et al
Table I
Basal characteristics in the groups of infants treated antenatally with betamethasone or dexamethasone
Physician-initiated cases Preeclampsia Fetal growth restriction Abruptio placentae Placenta praevia Chorioamnionitis Premature membrane rupture No !48 h 2-5 d O5 d Ritodrine No !48 h 2-5 d O5 d OOverall Cesarean section Corticosteroid dose 24 mg (complete course) 48 mg (1 additional course) 72 mg (2 additional courses) O72 mg (O2 additional courses) Surfactant use
Dexamethasone (n = 63)
Betamethasone (n = 138)
n (%)
n (%)
P value*
32 9 20 4 2 5
(47%) (14.3%) (31.7%) (6.3%) (3.2%) (7.9%)
85 18 50 12 4 9
(61.6%) (13%) (36.2%) (8.7%) (2.9%) (6.5%)
.2 .99 .65 O.999 O.999 .95
47 8 3 5
(74.6%) (12.7%) (4.8%) (7.9%)
115 23 5 8
(83.3%) (16.7%) (3.6%) (5.8%)
.5
35 1 9 18 28 44
(55.6%) (1.6%) (14.3%) (28.6%) (44.4%) (69.8%)
98 3 9 28 40 112
(71%) (2.2%) (6.5%) (20.3%) (29%) (81.2%)
.042 .047 .12
30 5 12 16 24
(47.6%) (7.9%) (19%) (25.4%) (38.1%)
101 10 16 11 48
(73.2%) (7.2%) (11.6%) (8%) (34.8%)
!.001 .77
* As obtained by c2, Fisher exact test or c2 for trend where appropriate.
birth weight ratio and surfactant administration, the OR of neurodevelopmental abnormalities at 2 years of corrected age was 2.56 (95% CI = 1.14-5.74) in untreated as opposed to infants treated antenatally with corticosteroids. No other significant differences in the neonatal outcome between the 2 groups could be detected. In the subsequent analysis, we sought to determine whether increasing exposure to multiple corticosteroids doses influenced infant outcomes. In univariate analysis, there was a significant linear trend relating increasing exposure to corticosteroids to prevalence of leukomalacia either expressed as prolonged echogenicity (c2 for trend = 4.84, P = .028) or as a whole (c2 for trend = 5.6, P = .018). In addition, the prevalence of infant neurodevelopmental abnormalities (from minor to severe) was significantly associated an increasing exposure to corticosteroids (c2 for trend = 3.89, P = .049). To perform a reliable multivariate analysis of trend, for each neonatal outcome, P values for trend were computed by logistic regression equations, including gestational age, birth weight ratio, multiple corticosteroids courses, ritodrine tocolysis, and surfactant use as single terms. The results of this analysis are reported in Table III. Increasing exposure to steroids was associated with a significantly increased risk of leukomalacia and neuro-
developmental abnormalities at 2 years of corrected age (adjusted P for trend = .011 and .038, respectively). Subsequently, we evaluated whether the increased risk of leukomalacia and 2-year infant neurodevelopmental abnormalities associated with increasing exposure to corticosteroids could be attributable to dexamethasone or betamethasone. The results of this stratified analysis are reported in Table IV. The risk of leukomalacia and neurodevelopmental abnormalities associated with exposure to multiple steroid dosages was almost entirely confined to the dexamethasone group. Multivariate study of interaction by logistic models confirmed the results of Table IV. In fact, after adjustment for gestational age, birth weight ratio, surfactant use, and ritodrine tocolysis, interaction terms between type of steroid and multiple exposure were significantly associated with the risks of leukomalacia and 2-year neurodevelopmental abnormalities (adjusted P value for interaction = .04 and .06, respectively).
Comment The potential limitations of the current study are related to the observational nature of the investigation and to
221
Spinillo et al
Table II ORS and 95% CI for various neonatal outcomes in the groups of infants treated antenatally with betamethasone or dexamethasone Dexamethasone (n = 63) Neonatal death Respiratory distress syndrome Assisted ventilation Acidosis at birth Acidosis first 24 h Convulsions Sepsis Intraventricular hemorrhage No Grade I-II Grade III-IV Overall Periventricular leukomalacia No Prolonged echogenicity Cystic leukomalacia Overall Neurologic outcome Normal (n = 134) Minor impairment (n = 26) Moderate-severe impairment (n = 12) Overall (n = 38)
Betamethasone (n = 138)
OR (95% CI)*
4 19 22 19 26 2 5
(6.3%) (30.2%) (34.9%) (30.2%) (41.3%) (3.2%) (7.9%)
7 43 41 22 27 2 9
(5.1%) (31.2%) (29.7%) (15.9%) (19.6%) (1.5%) (6.5%)
2.24 1.05 2.06 2.54 3.73 2.25 2.08
(0.47-10.9) (0.51-2.15) (0.97-4.37) (1.19-5.44) (1.82-7.67) (0.27-19.0) (0.65-6.67)
56 4 3 7
(88.9%) (6.3%) (4.8%) (11.1%)
131 4 3 7
(94.9%) (2.9%) (2.2%) (5.1%)
Referent 2.13 (0.54-8.4) 2.67 (0.46-15.46) 2.45 (0.77-7.80)
36 24 3 27
(57.1%) (38.1%) (4.8%) (42.9%)
101 33 4 37
(73.2%) (23.9%) (2.9%) (26.8%)
Referent 2.53 (1.13-5.65) 3.11 (0.57-16.78) 2.25 (1.07-4.74)
39 11 6 17
(69.6%) (19.6%) (10.7%) (30.4%)
95 15 6 21
(81.9%) (12.9%) (5.2%) (18.1%)
Referent 1.95 (0.73-5.18) 2.67 (0.79-9.02) 2.13 (0.94-4.84)
* As obtained by logistic regression equations containing neonatal outcome as dependent variable and gestational age, birth weight ratio (single terms), surfactant use (yes,no) multiple steroid dosages exposure (yes,no), and ritodrine tocolysis (yes,no) as independent variables.
the fact that the exposure to multiple steroid dosages was not uniform across betamethasone and dexamethasone categories. In particular, patients receiving dexamethasone were more likely to receive multiple courses and were also more likely to receive long-term ritodrine tocolysis. It is likely that the preference of some attending obstetricians could have been influenced by several European studies17,18 showing that multiple antenatal dexamethasone courses have a less profound effect on fetal heart rate variability compared with betamethasone. We sought to overcome these limitations by using logistic regression analysis to correct for confounders. Compared with betamethasone, dexamethasone administration was associated with an increased risk for neonatal acidosis both at birth and in the first 24 hours of life. The increased risk persisted after correction by logistic regression for the confounding effect of gestational age, birth weight ratio, ritodrine use, surfactant use, and multiple steroid dose exposure suggesting a direct dexamethasone causal role. Animal studies19 have demonstrated that betamethasone is much more efficient than dexamethasone in the process of maturational changes of immature lung. On the other hand, compared with betamethasone, repetitive doses of dexamethasone have been shown to increase the risk of neonatal death and suboptimal intrauterine growth in
mouse pups.19 The reduced rate of acidosis among preterm infants treated antenatally with betamethasone could reflect a better adaptative response of immature neonatal lung after this kind of treatment. Other literature data suggest that antenatal betamethasone is more effective than dexamethasone in the prevention of neonatal complications.20 For example, separate metaanalyses of the data in the Cochrane review show that only betamethasone rather than dexamethasone reduces neonatal mortality with statistical significance.21 Regarding the relationship between antenatal steroid treatment and subsequent neonatal periventricular leukomalacia, several studies22,23 have found that antenatal betamethasone exposure significantly reduce the risk of intraventricular hemorrhage and periventricular leukomalacia. In a retrospective analysis of 833 very preterm infants, Baud et al22 found that antenatal exposure to betamethasone, but not dexamethasone, decreased significantly the risk for periventricular leukomalacia. Our study confirms this finding suggesting that repeated antenatal courses of dexamethasone, as opposed to betamethasone, could be associated with an increased risk of periventricular leukomalacia. The relationship between antenatal exposure to multiple corticosteroid courses and subsequent long-term neonatal outcome is controversial.20 Animal studies
222 Table III exposure
Spinillo et al Analysis of linear trends in the prevalences of adverse neonatal events associated with increasing antenatal corticosteroid 24 mg (n = 131)
Neonatal death Respiratory distress Assisted ventilation Acidosis first 24 h Convulsions Sepsis Intraventricular hemorrhage No Grade I-II Grade III-IV Overall Periventricular leukomalacia No Prolonged echogenicity Cystic Overall Neurologic outcome Normal Minor impairment Moderate-severe impairment Overall
48 mg (n = 15)
72 mg (n = 28)
O 72 mg (n = 27)
P for trend*
9 38 43 35 1 10
d 6 6 3 d 1
1 13 6 6 d d
1 5 8 9 2 3
.4 .85 .49 .64 .11 .91
123 5 3 8
14 1 d 1
26 d 2 2
24 2 1 3
.33
97 31 3 34
9 5 1 6
16 10 2 12
15 11 1 12
.011
91 14 6 20
11 2 1 3
17 4 3 7
15 6 2 8
.038
* As obtained by logistic regression equations containing neonatal outcome as dependent variable and gestational age, birth weight ratio, multiple steroid dosages exposure (single terms), surfactant use (yes,no), and ritodrine tocolysis (yes,no) as independent variables. Steroids dosage was coded as 0 = 24 mg; 1 = 48 mg; 2 = 72 mg; and 3 = O 72 mg and included as single term.
Table IV used
Prevalence of leukomalacia and overall neurodevelopmental abnormalities according to type of corticosteroid and dosage
Leukomalacia No Yes
Single dose
Multiple doses
Dexa (n = 30) Beta (n = 101) OR (95% CI)
Dexa (n = 33) Beta (n = 37) OR (95% CI)
22 8 (n = 28)
Neurodevelopmental abnormalities No 23 Yes 5
75 26 (n = 81)
Referent 14 1.05 (0.38-2.87) 19 (n = 28)
26 11 (n = 35)
Referent 3.21 (1.07-9.77)
66 15
Referent 16 0.96 (0.27-3.25) 12
29 6
Referent 3.63 (1.03-13.6)
Dexa, Dexamethasone; Beta, betamethasone.
have shown that repeated courses of betamethasone are associated with significant brain growth restriction and severe neurodevelopmental effects.8 In addition, wellcontrolled pediatric studies have shown that postnatal dexamethasone administration is significantly associated with an increased risk of poor neurodevelopmental outcome.24 Human studies have failed to demonstrate a significant effect of antenatal steroids administration on postmortem brain weight of preterm infants,25 but postnatal magnetic resonance indices of infant brain maturation suggest a potential negative effect of multiple antenatal steroid administrations.26
Published data on the infant neurodevelopmental outcome after antenatal exposure to multiple steroid dosages are scant. In the study of French et al10 multiple antenatal betamethasone courses were associated with a progressive reduction of head circumferences at birth. At the 3-year follow-up, the neurodevelopmental outcome of 43 infants exposed to multiple antenatal betamethasone was similar to controls. Hasbargen et al11 failed to ascertain adverse long-term neurodevelopmental effects up to the age of 4 years in 28 infants exposed antenatally to more than 5 courses of betamethasone. In our study, antenatal exposure to multiple steroid cour-
223
Spinillo et al ses, especially dexamethasone, was associated with an increased risk of 2-year infant neurodevelopmental abnormalities. Given the limited statistical power of the current investigation, to avoid the possibility of a type I error, we acknowledge that our results need to be confirmed on larger studies. However, it is clear that multiple antenatal steroid dosages did not confer any significant advantage in the prevention of both shortand long-term infant morbidity. Regarding the role of dexamethasone as opposed to betamethasone, several studies suggest that dexamethasone could cause specific neurologic damages through different pathways. Animal studies suggest that sulphite ions present in dexamethasone vehicles (1 mg NaS2O5 for 5 mg dexamethasone) in combination with peroxynitrite could impair neural cell viability.27 Acute administration of dexamethasone induces dose-dependent apoptosis and extensive sublethal injury to specific subpopulations of striatal and hippocampal neurons in rats; this toxic effect is remarkably specific to this type of steroid.28 In addition, a recent study on premature infants indicates that postnatal dexamethasone administration for chronic lung disease causes an impairment in brain growth, principally affecting cerebral cortical grey matter.29 On the basis of these data and according to Whitelaw et al,20 we suggest that obstetricians using multiple antenatal dexamethasone dosages should consider that this approach could have adverse effects on the developing brain of premature fetuses. In conclusion, the current study suggests that antenatal exposure to multiple steroid courses of dexamethasone could have adverse effects on 2-year infant neurodevelopmental outcome. In agreement with other authors, we confirm that multiple courses of corticosteroids, particularly dexamethasone, should not be routinely prescribed to women at risk of preterm delivery.
7.
8.
9.
10.
11.
12.
13.
14. 15.
16. 17.
18.
19.
20.
References 21. 1. Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treament for prevention of the respiratory distress syndrome in premature infants. Pediatrics 1972;50:515-25. 2. Papageorgiou AN, Desgranges MF, Masson M, Colle E, Shatz R, Gelfand MM. The antenatal use of betamethasone in the prevention of respiratory distress syndrome: a controlled double-blind study. Pediatrics 1979;63:73-9. 3. Crowley P, Chalmers I, Keirse MJ. The effects of corticosteroid administration before preterm delivery: an overview of the evidence from controlled trials. BJOG 1990;97:11-25. 4. National Institutes of Health. Report of the consensus development conference on the effect of corticosteroids for fetal maturation on perinatal outcome. Bethesda, MD: National Institute of Child Health and Human Development; November 1994. NIH publication no. 95-3784. 5. National Institutes of Health. Antenatal cortiocsteroids revisited: Repeat courses. NIH Consensus Statement 2000;17:1-18. 6. Brockelehurst P, Gates S, McKenzie-McHarg K, Alfrevic Z, Chamberlain G. Are we prescribing multiple courses of antenatal
22.
23.
24.
25. 26.
corticosteroids? A survey of practice in the UK. BJOG 1998;106: 977-9. Guinn DA, Atkinson MW, Sullivan L, Lee M, MacGregor S, Parilla BV, et al. Single vs weekly courses of antenatal corticosteroids for women at risk of preterm delivery. JAMA 2001;286: 1581-7. Huang WL, Harper CG, Evans SF, Newnham JP, Dunlop SA. Repeated prenatal corticosteroid administration delays astrocyte and capillary tight junction maturation in fetal sheep. Int J Dev Neurosci 2000;19:487-93. Hasbargen U, Reber D, Versmold H, Schulze A. Growth and development of children to 4 years of age after repeated antenatal steroid administration. Eur J Pediatr 2001;160:552-5. French NP, Hagan R, Evans SF, Godfrey M, Newnham JP. Repeated antenatal corticosteroids: size at birth and subsequent development. Am J Obstet Gynecol 1999;180:114-21. Davey DA, McGillivray IM. The classification and definition of the hypertensive disorders of pregnancy. Am J Obstet Gynecol 1988;158:892-8. Gibbs RS, Duff P. Progress in pathogenesis and management of clinical intraamniotic infection. Am J Obstet Gynecol 1991;164: 1317-26. Papile L, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weight less than 1500 gm. J Pediatr 1978;92:529-34. Amiel-Tison C, Grenier A. Evaluation neurologique du nouveau-ne et du nourisson. Paris: Masson; 1980. de Vries LS, Dubowitz LM, Dubowitz V, Kaiser A, Lary S, Silverman M, et al. Predictive value of cranial ultrasound in the newborn baby: a reappraisal. Lancet 1985;2:137-40. Hosmer DW, Lemeshow S. Applied logistic regression. New York: John Wiley; 1989. p. 720e4. Senat MV, Minoui V, Multon O, Fernandez H, Frydman R, Ville Y. Effect of dexamethasone and betamethasone on fetal heart rate variability in preterm labour: a randomized study. BJOG 1998;105: 749-55. Multon O, Senat MV, Minoui S, Hue MV, Frydman R, Ville Y. Effect of antenatal betamethasone and dexamethasone on fetal heart rate variability in growth retarded fetuses. Fetal Diagn Ther 1997;12:170-7. Ozdemir H, Guvenal T, Cetin M, Kaya T, Cetin A. A placebocontrolled comparison of effects of repetitive doses of betamethasone and dexamethasone on lung maturation and lung, liver, and body weights of mouse pups. Pediatr Res 2003;53:98-103. Whitelaw A, Thoresen M. Antenatal steroids and the developing brain. Arch Dis Child Fetal Neonatal 2002;83:F154-7. Crowley P. Prophylactic corticosteroids for preterm delivery. In: The Cochrane library. Issue 4. Oxford: Update Software; 1999. Baud O, Foix-L’Helias L, Kaminski M, Audibert F, Jarreau PH, Papiernik E, et al. Antenatal glucocorticoid treatment and cystic periventricular leukomalacia in very premature infants. N Engl J Med 1999;341:1190-6. Canterino JC, Verma U, Visintainer PF, Elimian A, Klein SA, Tejani N. Antenatal steroids and neonatal periventricular leukomalacia. Obstet Gynecol 2001;97:135-9. O’Shea TM, Kothadia JM, Klinepeter KL, Goldstein DJ, Jackson BG, Weaver RG 3rd, et al. Randomized placebo-controlled trial of a 42-day tapering course of dexamethasone to reduce the duration of ventilator dependency in very low birthweight infants: outcome of study participants at 1-year adjusted age. Pediatrics 1999;104: 15-21. Murphy DJ. Effect of antenatal corticosteroids on postmortem brain weight of preterm babies. Early Hum Dev 2001;63:113-22. Modi N, Lewis H, Al-Naqeeb N, Ajayi-Obe M, Dore CJ, Rutheford M. The effects of repeated antenatal glucocorticoid therapy on the developing brain. Pediatr Res 2001;50:563-4.
224 27. Reist M, Marshall KA, Jenner P, Halliwell B. Toxic effects of sulphite in combination with peroxynitrite on neuronal cells. J Neurochem 1998;71:2431-8. 28. Haynes LE, Griffiths MR, Hyde RE, Barber DJ, Mitchell IJ. Dexamethasone induces limited apoptosis and extensive sublethal damage to specific subregions of the striatum and
Spinillo et al hippocampus: implications for mood disorders. Neuroscience 2001;104:57-69. 29. Murphy BP, Inder TE, Huppi PS, Warfield S, Zientara GP, Kikinis R, et al. Impaired cerebral cortex gray matter growth after treatment with dexamethasone for neonatal chronic lung disease. Pediatrics 2001;107:217-21.
www.elsevier.com/locate/ajog
Two-year infant neurodevelopmental outcome after single or multiple antenatal courses of corticosteroids to prevent complications of prematurity Arsenio Spinillo,a,* Franco Viazzo,a Rossella Colleoni,a Alberto Chiara,b Rosa Maria Cerbo,b Elisa Fazzic Departments of Obstetrics and Gynecology,a Neonatal Intensive Care,b and Child Neuropsychiatry,c University of Pavia, Istituto di Ricovero e Cura a Carattere Scientifico Policlinico S. Matteo ed Istituto di Ricovero e Cura a Carattere Scientifico C. Mondino, Pavia, Italy Received for publication October 14, 2003; revised December 10, 2003; accepted December 22, 2003
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– KEY WORDS Betamethasone Dexamethasone Periventricular leukomalacia Prematurity
Objective: This study was undertaken to evaluate the effect of exposure to multiple antenatal steroid courses on short-term neonatal morbidity and 2-year infant neurodevelopmental outcome. Study design: This was a prospective observational study of 201 preterm singleton infants who received 1 or more courses of corticosteroids to prevent complications of prematurity and were delivered between 24 and 34 weeks’ gestation at a single institution. Neurodevelopmental outcome of the infants was evaluated at 2 years corrected age. Logistic regression analysis was used to perform multivariate analyses of associations and trends. Results: One hundred thirty-eight subjects (68.7%) received at least 1 complete course of betamethasone, whereas 63 (31.3%) patients were treated with dexamethasone. The prevalence of multiple steroid doses exposure was 26.8% (37/138) in betamethasone and 52.4% (33/63) in dexamethasone group. The prevalence of infant leukomalacia, including both prolonged echogenicity and cystic leukomalacia, was 25.9% (34/131) after a complete corticosteroid course, 40% (6/15) after 1, 42.3% (12/28) after 2, and 44.4% (12/27) after more than 2 additional courses, respectively (adjusted P for trend = .011). In the same categories of steroid exposure, the corresponding prevalences of 2-year infant neurodevelopmental abnormalities were 18% (20/111), 21.4% (3/14), 29.2% (7/24), and 34.8% (8/23), respectively (adjusted P for trend = .038). Multivariate study of first grade interaction suggested that the risk of leukomalacia and 2-year infant neurodevelopmental abnormalities associated with multiple doses exposure was confined to dexamethasone. In fact, compared with betamethasone, exposure to multiple doses of dexamethasone was associated with an increased risk of leukomalacia (19/33 compared with 11/37; odds ratio [OR] = 3.21, 95% CI = 1.07-9.77) and overall 2-year infant neurodevelopmental abnormalities (12/28 compared with 6/35; OR = 3.63, 95% CI = 1.03-13.58).
* Reprint requests: Arsenio Spinillo, MD, Department of Obstetrics and Gynecology, University of Pavia. Policlinico S. Matteo, P. le Golgi,19, 27100 Pavia, Italy E-mail: [email protected] 0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ajog.2003.12.023
218
Spinillo et al Conclusion: In this study, multiple antenatal courses of dexamethasone but not betamethasone were associated with an increased risk of leukomalacia and 2-year infant neurodevelopmental abnormalities. Ó 2004 Elsevier Inc. All rights reserved.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– A large number of controlled, observational studies and metanalyses support the use of antenatal corticosteroids to prevent neonatal complications of prematurity in women at risk of preterm delivery.1-3 A consensus panel of the National Institutes of Health in 1994 concluded that antenatal betamethasone or dexamethasone reduces the risk of neonatal death, respiratory distress syndrome, and intraventricular hemorrhage when administered up to 7 days before delivery.4 The same panel concluded that data were insufficient to establish a clear benefit beyond 7 days. A subsequent reconvening of the Consensus Conference cautioned against the use of repeated antenatal corticosteroids outside a randomized control trial.5 In recent years, it has been a common practice among obstetricians to prescribe weekly antenatal steroid courses to women undelivered after the first dose.6 In a recent randomized study, weekly courses of antenatal corticosteroids did not reduce short-term composite neonatal morbidity (perinatal death, respiratory distress syndrome, intraventricular hemorrhage, and periventricular leukomalacia) compared with a single course of treatment.7 In addition, animal data suggest that repeated antenatal steroid doses could interfere with the growth and development of the immature brain.8 Data on the effect of weekly antenatal corticosteroid courses on infant neurodevelopmental outcome are sparse and mainly related to antenatal betamethasone rather than dexamethasone administration.9,10 The purpose of this study was to evaluate the 2-year infant neurodevelopmental outcome in a cohort of women treated antenatally with single or repeated courses of either betamethasone or dexamethasone.
Methods This is a prospective observational study of all pregnant women delivered of a singleton live fetus between 24 and 34 weeks of gestation at a third level regional perinatal center between January 1, 1997, and December 3, 2000, who had received 1 or more antenatal courses of corticosteroid for the prevention of the complications of prematurity. Of a total of 315 singleton pregnancies delivered in the period examined; 16 severely malformed fetuses, 10 pregnancies with uncertain gestational age, 63 pregnancies delivered without corticosteroids treatment, and 25 with incomplete steroid administration (single dose) were excluded from the study. The remaining 201 women who received at least 1 complete antena-
tal steroid course constituted the study group. In these patients antenatal steroids were routinely administered either in the form of two 12-mg intramuscular doses of betamethasone, 24 hours apart, or dexamethasone, 6-mg intramuscular every 12 hours for 4 doses depending on the attending obstetrician. Subsequently, if the patient had not delivered, weekly repetitions of betamethasone or dexamethasone treatments were carried out at the discretion of the attending physician. Maternal data and clinical variables were collected prospectively and stored in a computer database. Preeclampsia was defined according to standard criteria.11 Fetal growth restriction was diagnosed when abdominal circumference ultrasonographic measurements decreased to below the 10th percentile of a standard growth curve for our population; the abnormal measurement of abdominal circumference was confirmed on at least 2 ultrasonographic examinations, 2 to 4 weeks apart, after the standard 18 to 22 weeks’ sonogram. To evaluate the severity of growth restriction, we also computed the birth weight ratio of each infant at any given gestational age. The birth weight ratio is the ratio between the observed and the expected birth weight, where expected birth weight is the 50th percentile of birth weight of our population corrected for the sex of the infant. The diagnosis of preterm premature rupture of membranes was based on clinical assessment, measurement of vaginal pH and ultrasonographic findings. Ritodrine was used to achieve tocolysis during the study period. The clinical diagnosis of chorioamnionitis was based on maternal temperature greater than 37.8(C plus 2 of the followings: (1) maternal tachycardia (O100 beats/min); (2) baseline fetal hearth rate greater than 160 beats/min; (3) positive endocervical or high-vaginal swab culture; (4) maternal white blood cell count greater than 18.000/ mm3; and (5) foul-smelling amniotic fluid. These criteria are similar to those recommended by Gibbs and Duff.12 After delivery all newborn infants were admitted to the neonatal intensive care unit of our department. A standard echoencephalogram was obtained within 24 hours of admission from all infants. Subsequently, serial cranial ultrasound was carried out every 48 to 72 hours during the first 2 weeks of life and weekly thereafter until they were discharged. Cranial ultrasounds were performed by 2 neonatologists (A.C. and R.M.C.) unaware of the type and dosages of antenatal steroid used. Infant intracranial hemorrhage was graded into 4 categories according to Papile et al.13 Periventricular leukomalacia included both infants with frank cystic de-
Spinillo et al generation of periventricular white matter and infants with persistently (more than a week) increased echodensity in the periventricular regions but not undergoing cystic degeneration. Neurodevelopmental evaluation of the infants was carried out by a child neuropsychiatrist not involved in the intensive care of the infants and unaware of maternal history. Examinations were carried out at discharge from hospital and at 3, 6, 12, and 24 months of corrected age. Neurologic evaluation of the newborn infants was based on the methods of AmielTison and Grenier.14 The Bayley scales of infant development were used to assess cognitive development (Mental Developmental Index [MDI]) at 12 to 24 months. Infants were grouped into 4 categories according to their final examination: (1) unimpaired = normal neurologic status and Bayley MDI O84; (2) minor impairment = abnormalities of tone or reflexes but functionally normal or borderline (71-84) Bayley MDI; (3) moderate impairment = spastic diplegia or hemiplegia with a Bayley MDI R71; and (4) severe impairment = spastic tetraplegia and/or severe mental retardation (Bayley MDI !71). This classification is similar to that proposed by De Vries et al.15 Among the 190 surviving infants treated antenatally with corticosteroids, 172 (90.5%) were followed-up throughout the study program. The 18 infants lost to follow-up had been judged neurologically normal at discharge from hospital. Of the 56 surviving infants who did not receive antenatal corticosteroids treatment, 51 (91.1%) were followed-up at 2 years of corrected age. Statistical analysis was carried out by using the Mann-Whitney U test or the c2 test to compare continuous and categorical variables, respectively. The c2 for trend was used to evaluate linear trends across ordered categories. Continuous variables are expressed as median and interquartile range when skewed or as mean and SD when normally distributed. Unconditional logistic regression was used to test for association between variables while adjusting for potential confounders and to test for linear trends and first-degree interaction between variables.16 Because this was not a randomized study, we did not compute a preliminary sample size. However, assuming a 20% prevalence of any given outcome (eg, overall intraventricular hemorrhage or leukomalacia) we had a 83% power to detect an odds ratio (OR) of 2 or more at the usual a values of .05.
Results The main antenatal characteristics of the studied population are reported in Table I. One hundred thirty-eight subjects (68.7%) received at least 1 complete course of betamethasone, whereas 63 (31.3%) patients were treated with dexamethasone. The median gestational age and birth weight were 31 weeks (interquartile
219 range = 4) and 1490 g (interquartile range = 670) in the betamethasone group and 31 weeks (interquartile range = 4) and 1456 g (interquartile range = 797) in the dexamethasone group (P = .88 and P = .25 by Mann-Whitney test compared with betamethasone, respectively). The median gestational age and birth weight of the 63 infants delivered without any antenatal steroid treatment mostly because of an emergency delivery and not presented in Table I were 33 weeks (interquartile range = 3) and 1759 g (interquartile range = 780) (P!.01 compared with treated infants for both comparisons). The mean values of birth weight ratio were 0.892 G 0.19 in the group of infants treated antenatally with betamethasone and 0.907 G 0.26 in those treated with dexamethasone (P = .94 by Mann-Whitney test). The corresponding value of birth weight ratio among untreated infants was 0.91 G 0.21. Patients who received dexamethasone were more likely to receive long-term tocolysis with ritodrine (c2 for trend = 4.14, P = .042) and multiple doses (c2 for trend = 15.7, P!.01). In addition, the median time interval from last steroid administration and birth was 3.2 days (range 1 to 7) in mothers receiving multiple and 5.7 days (range 1 to 28) in those receiving single doses (P!.001 by Mann-Whitney test). In Table II are reported the multivariate ORs of neonatal outcome variables adjusted by gestational age, birth weight ratio, ritodrine use, surfactant use, and multiple steroids dosages among infants born to mothers treated with dexamethasone as opposed to betamethasone. The occurrence of neonatal acidosis (arterial pH !7.2) both at birth and in the first 24 hours of life was significantly higher among infants delivered after antenatal dexamethasone administration compared with betamethasone treatment. Transient periventricular echogenicity, disappearing in the first week of life was recorded in 4 infants (6.3%) treated antenatally with dexamethasone and in 11 (8%) with betamethasone, respectively. After adjustment for potential confounders, the prevalence of periventricular leukomalacia was significantly higher among infants treated antenatally with dexamethasone as opposed to betamethasone (OR = 2.25, 95% CI = 1.07-4.74). The relative rates of tetraplegia and emiplegia-diplegia were 3.5 % (5/141) and 2.8% (4/141) in the betamethasone group and 3.6% (2/55) and 7.3% (4/55) in the dexamethasone group, respectively. In the group of 63 infants delivered without any steroid treatment, the prevalences of respiratory distress syndrome and neonatal death were 41.3% (26 of 63; OR = 1.58, 95% CI = 0.84-2.93 compared with treated infants) and 11.1% (7 of 63; OR = 2.16, 95% CI = 0.726.38 compared with treated infants). In addition, the prevalences of moderate-severe impairment and minor neurodevelopmental abnormalities were 7.8% (4/51) and 31.4% (16/51) (c2, 2 df = 7.12, P = .028 in comparison to treated infants), respectively. In logistic regression models, after adjustment for gestational age,
220
Spinillo et al
Table I
Basal characteristics in the groups of infants treated antenatally with betamethasone or dexamethasone
Physician-initiated cases Preeclampsia Fetal growth restriction Abruptio placentae Placenta praevia Chorioamnionitis Premature membrane rupture No !48 h 2-5 d O5 d Ritodrine No !48 h 2-5 d O5 d OOverall Cesarean section Corticosteroid dose 24 mg (complete course) 48 mg (1 additional course) 72 mg (2 additional courses) O72 mg (O2 additional courses) Surfactant use
Dexamethasone (n = 63)
Betamethasone (n = 138)
n (%)
n (%)
P value*
32 9 20 4 2 5
(47%) (14.3%) (31.7%) (6.3%) (3.2%) (7.9%)
85 18 50 12 4 9
(61.6%) (13%) (36.2%) (8.7%) (2.9%) (6.5%)
.2 .99 .65 O.999 O.999 .95
47 8 3 5
(74.6%) (12.7%) (4.8%) (7.9%)
115 23 5 8
(83.3%) (16.7%) (3.6%) (5.8%)
.5
35 1 9 18 28 44
(55.6%) (1.6%) (14.3%) (28.6%) (44.4%) (69.8%)
98 3 9 28 40 112
(71%) (2.2%) (6.5%) (20.3%) (29%) (81.2%)
.042 .047 .12
30 5 12 16 24
(47.6%) (7.9%) (19%) (25.4%) (38.1%)
101 10 16 11 48
(73.2%) (7.2%) (11.6%) (8%) (34.8%)
!.001 .77
* As obtained by c2, Fisher exact test or c2 for trend where appropriate.
birth weight ratio and surfactant administration, the OR of neurodevelopmental abnormalities at 2 years of corrected age was 2.56 (95% CI = 1.14-5.74) in untreated as opposed to infants treated antenatally with corticosteroids. No other significant differences in the neonatal outcome between the 2 groups could be detected. In the subsequent analysis, we sought to determine whether increasing exposure to multiple corticosteroids doses influenced infant outcomes. In univariate analysis, there was a significant linear trend relating increasing exposure to corticosteroids to prevalence of leukomalacia either expressed as prolonged echogenicity (c2 for trend = 4.84, P = .028) or as a whole (c2 for trend = 5.6, P = .018). In addition, the prevalence of infant neurodevelopmental abnormalities (from minor to severe) was significantly associated an increasing exposure to corticosteroids (c2 for trend = 3.89, P = .049). To perform a reliable multivariate analysis of trend, for each neonatal outcome, P values for trend were computed by logistic regression equations, including gestational age, birth weight ratio, multiple corticosteroids courses, ritodrine tocolysis, and surfactant use as single terms. The results of this analysis are reported in Table III. Increasing exposure to steroids was associated with a significantly increased risk of leukomalacia and neuro-
developmental abnormalities at 2 years of corrected age (adjusted P for trend = .011 and .038, respectively). Subsequently, we evaluated whether the increased risk of leukomalacia and 2-year infant neurodevelopmental abnormalities associated with increasing exposure to corticosteroids could be attributable to dexamethasone or betamethasone. The results of this stratified analysis are reported in Table IV. The risk of leukomalacia and neurodevelopmental abnormalities associated with exposure to multiple steroid dosages was almost entirely confined to the dexamethasone group. Multivariate study of interaction by logistic models confirmed the results of Table IV. In fact, after adjustment for gestational age, birth weight ratio, surfactant use, and ritodrine tocolysis, interaction terms between type of steroid and multiple exposure were significantly associated with the risks of leukomalacia and 2-year neurodevelopmental abnormalities (adjusted P value for interaction = .04 and .06, respectively).
Comment The potential limitations of the current study are related to the observational nature of the investigation and to
221
Spinillo et al
Table II ORS and 95% CI for various neonatal outcomes in the groups of infants treated antenatally with betamethasone or dexamethasone Dexamethasone (n = 63) Neonatal death Respiratory distress syndrome Assisted ventilation Acidosis at birth Acidosis first 24 h Convulsions Sepsis Intraventricular hemorrhage No Grade I-II Grade III-IV Overall Periventricular leukomalacia No Prolonged echogenicity Cystic leukomalacia Overall Neurologic outcome Normal (n = 134) Minor impairment (n = 26) Moderate-severe impairment (n = 12) Overall (n = 38)
Betamethasone (n = 138)
OR (95% CI)*
4 19 22 19 26 2 5
(6.3%) (30.2%) (34.9%) (30.2%) (41.3%) (3.2%) (7.9%)
7 43 41 22 27 2 9
(5.1%) (31.2%) (29.7%) (15.9%) (19.6%) (1.5%) (6.5%)
2.24 1.05 2.06 2.54 3.73 2.25 2.08
(0.47-10.9) (0.51-2.15) (0.97-4.37) (1.19-5.44) (1.82-7.67) (0.27-19.0) (0.65-6.67)
56 4 3 7
(88.9%) (6.3%) (4.8%) (11.1%)
131 4 3 7
(94.9%) (2.9%) (2.2%) (5.1%)
Referent 2.13 (0.54-8.4) 2.67 (0.46-15.46) 2.45 (0.77-7.80)
36 24 3 27
(57.1%) (38.1%) (4.8%) (42.9%)
101 33 4 37
(73.2%) (23.9%) (2.9%) (26.8%)
Referent 2.53 (1.13-5.65) 3.11 (0.57-16.78) 2.25 (1.07-4.74)
39 11 6 17
(69.6%) (19.6%) (10.7%) (30.4%)
95 15 6 21
(81.9%) (12.9%) (5.2%) (18.1%)
Referent 1.95 (0.73-5.18) 2.67 (0.79-9.02) 2.13 (0.94-4.84)
* As obtained by logistic regression equations containing neonatal outcome as dependent variable and gestational age, birth weight ratio (single terms), surfactant use (yes,no) multiple steroid dosages exposure (yes,no), and ritodrine tocolysis (yes,no) as independent variables.
the fact that the exposure to multiple steroid dosages was not uniform across betamethasone and dexamethasone categories. In particular, patients receiving dexamethasone were more likely to receive multiple courses and were also more likely to receive long-term ritodrine tocolysis. It is likely that the preference of some attending obstetricians could have been influenced by several European studies17,18 showing that multiple antenatal dexamethasone courses have a less profound effect on fetal heart rate variability compared with betamethasone. We sought to overcome these limitations by using logistic regression analysis to correct for confounders. Compared with betamethasone, dexamethasone administration was associated with an increased risk for neonatal acidosis both at birth and in the first 24 hours of life. The increased risk persisted after correction by logistic regression for the confounding effect of gestational age, birth weight ratio, ritodrine use, surfactant use, and multiple steroid dose exposure suggesting a direct dexamethasone causal role. Animal studies19 have demonstrated that betamethasone is much more efficient than dexamethasone in the process of maturational changes of immature lung. On the other hand, compared with betamethasone, repetitive doses of dexamethasone have been shown to increase the risk of neonatal death and suboptimal intrauterine growth in
mouse pups.19 The reduced rate of acidosis among preterm infants treated antenatally with betamethasone could reflect a better adaptative response of immature neonatal lung after this kind of treatment. Other literature data suggest that antenatal betamethasone is more effective than dexamethasone in the prevention of neonatal complications.20 For example, separate metaanalyses of the data in the Cochrane review show that only betamethasone rather than dexamethasone reduces neonatal mortality with statistical significance.21 Regarding the relationship between antenatal steroid treatment and subsequent neonatal periventricular leukomalacia, several studies22,23 have found that antenatal betamethasone exposure significantly reduce the risk of intraventricular hemorrhage and periventricular leukomalacia. In a retrospective analysis of 833 very preterm infants, Baud et al22 found that antenatal exposure to betamethasone, but not dexamethasone, decreased significantly the risk for periventricular leukomalacia. Our study confirms this finding suggesting that repeated antenatal courses of dexamethasone, as opposed to betamethasone, could be associated with an increased risk of periventricular leukomalacia. The relationship between antenatal exposure to multiple corticosteroid courses and subsequent long-term neonatal outcome is controversial.20 Animal studies
222 Table III exposure
Spinillo et al Analysis of linear trends in the prevalences of adverse neonatal events associated with increasing antenatal corticosteroid 24 mg (n = 131)
Neonatal death Respiratory distress Assisted ventilation Acidosis first 24 h Convulsions Sepsis Intraventricular hemorrhage No Grade I-II Grade III-IV Overall Periventricular leukomalacia No Prolonged echogenicity Cystic Overall Neurologic outcome Normal Minor impairment Moderate-severe impairment Overall
48 mg (n = 15)
72 mg (n = 28)
O 72 mg (n = 27)
P for trend*
9 38 43 35 1 10
d 6 6 3 d 1
1 13 6 6 d d
1 5 8 9 2 3
.4 .85 .49 .64 .11 .91
123 5 3 8
14 1 d 1
26 d 2 2
24 2 1 3
.33
97 31 3 34
9 5 1 6
16 10 2 12
15 11 1 12
.011
91 14 6 20
11 2 1 3
17 4 3 7
15 6 2 8
.038
* As obtained by logistic regression equations containing neonatal outcome as dependent variable and gestational age, birth weight ratio, multiple steroid dosages exposure (single terms), surfactant use (yes,no), and ritodrine tocolysis (yes,no) as independent variables. Steroids dosage was coded as 0 = 24 mg; 1 = 48 mg; 2 = 72 mg; and 3 = O 72 mg and included as single term.
Table IV used
Prevalence of leukomalacia and overall neurodevelopmental abnormalities according to type of corticosteroid and dosage
Leukomalacia No Yes
Single dose
Multiple doses
Dexa (n = 30) Beta (n = 101) OR (95% CI)
Dexa (n = 33) Beta (n = 37) OR (95% CI)
22 8 (n = 28)
Neurodevelopmental abnormalities No 23 Yes 5
75 26 (n = 81)
Referent 14 1.05 (0.38-2.87) 19 (n = 28)
26 11 (n = 35)
Referent 3.21 (1.07-9.77)
66 15
Referent 16 0.96 (0.27-3.25) 12
29 6
Referent 3.63 (1.03-13.6)
Dexa, Dexamethasone; Beta, betamethasone.
have shown that repeated courses of betamethasone are associated with significant brain growth restriction and severe neurodevelopmental effects.8 In addition, wellcontrolled pediatric studies have shown that postnatal dexamethasone administration is significantly associated with an increased risk of poor neurodevelopmental outcome.24 Human studies have failed to demonstrate a significant effect of antenatal steroids administration on postmortem brain weight of preterm infants,25 but postnatal magnetic resonance indices of infant brain maturation suggest a potential negative effect of multiple antenatal steroid administrations.26
Published data on the infant neurodevelopmental outcome after antenatal exposure to multiple steroid dosages are scant. In the study of French et al10 multiple antenatal betamethasone courses were associated with a progressive reduction of head circumferences at birth. At the 3-year follow-up, the neurodevelopmental outcome of 43 infants exposed to multiple antenatal betamethasone was similar to controls. Hasbargen et al11 failed to ascertain adverse long-term neurodevelopmental effects up to the age of 4 years in 28 infants exposed antenatally to more than 5 courses of betamethasone. In our study, antenatal exposure to multiple steroid cour-
223
Spinillo et al ses, especially dexamethasone, was associated with an increased risk of 2-year infant neurodevelopmental abnormalities. Given the limited statistical power of the current investigation, to avoid the possibility of a type I error, we acknowledge that our results need to be confirmed on larger studies. However, it is clear that multiple antenatal steroid dosages did not confer any significant advantage in the prevention of both shortand long-term infant morbidity. Regarding the role of dexamethasone as opposed to betamethasone, several studies suggest that dexamethasone could cause specific neurologic damages through different pathways. Animal studies suggest that sulphite ions present in dexamethasone vehicles (1 mg NaS2O5 for 5 mg dexamethasone) in combination with peroxynitrite could impair neural cell viability.27 Acute administration of dexamethasone induces dose-dependent apoptosis and extensive sublethal injury to specific subpopulations of striatal and hippocampal neurons in rats; this toxic effect is remarkably specific to this type of steroid.28 In addition, a recent study on premature infants indicates that postnatal dexamethasone administration for chronic lung disease causes an impairment in brain growth, principally affecting cerebral cortical grey matter.29 On the basis of these data and according to Whitelaw et al,20 we suggest that obstetricians using multiple antenatal dexamethasone dosages should consider that this approach could have adverse effects on the developing brain of premature fetuses. In conclusion, the current study suggests that antenatal exposure to multiple steroid courses of dexamethasone could have adverse effects on 2-year infant neurodevelopmental outcome. In agreement with other authors, we confirm that multiple courses of corticosteroids, particularly dexamethasone, should not be routinely prescribed to women at risk of preterm delivery.
7.
8.
9.
10.
11.
12.
13.
14. 15.
16. 17.
18.
19.
20.
References 21. 1. Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treament for prevention of the respiratory distress syndrome in premature infants. Pediatrics 1972;50:515-25. 2. Papageorgiou AN, Desgranges MF, Masson M, Colle E, Shatz R, Gelfand MM. The antenatal use of betamethasone in the prevention of respiratory distress syndrome: a controlled double-blind study. Pediatrics 1979;63:73-9. 3. Crowley P, Chalmers I, Keirse MJ. The effects of corticosteroid administration before preterm delivery: an overview of the evidence from controlled trials. BJOG 1990;97:11-25. 4. National Institutes of Health. Report of the consensus development conference on the effect of corticosteroids for fetal maturation on perinatal outcome. Bethesda, MD: National Institute of Child Health and Human Development; November 1994. NIH publication no. 95-3784. 5. National Institutes of Health. Antenatal cortiocsteroids revisited: Repeat courses. NIH Consensus Statement 2000;17:1-18. 6. Brockelehurst P, Gates S, McKenzie-McHarg K, Alfrevic Z, Chamberlain G. Are we prescribing multiple courses of antenatal
22.
23.
24.
25. 26.
corticosteroids? A survey of practice in the UK. BJOG 1998;106: 977-9. Guinn DA, Atkinson MW, Sullivan L, Lee M, MacGregor S, Parilla BV, et al. Single vs weekly courses of antenatal corticosteroids for women at risk of preterm delivery. JAMA 2001;286: 1581-7. Huang WL, Harper CG, Evans SF, Newnham JP, Dunlop SA. Repeated prenatal corticosteroid administration delays astrocyte and capillary tight junction maturation in fetal sheep. Int J Dev Neurosci 2000;19:487-93. Hasbargen U, Reber D, Versmold H, Schulze A. Growth and development of children to 4 years of age after repeated antenatal steroid administration. Eur J Pediatr 2001;160:552-5. French NP, Hagan R, Evans SF, Godfrey M, Newnham JP. Repeated antenatal corticosteroids: size at birth and subsequent development. Am J Obstet Gynecol 1999;180:114-21. Davey DA, McGillivray IM. The classification and definition of the hypertensive disorders of pregnancy. Am J Obstet Gynecol 1988;158:892-8. Gibbs RS, Duff P. Progress in pathogenesis and management of clinical intraamniotic infection. Am J Obstet Gynecol 1991;164: 1317-26. Papile L, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weight less than 1500 gm. J Pediatr 1978;92:529-34. Amiel-Tison C, Grenier A. Evaluation neurologique du nouveau-ne et du nourisson. Paris: Masson; 1980. de Vries LS, Dubowitz LM, Dubowitz V, Kaiser A, Lary S, Silverman M, et al. Predictive value of cranial ultrasound in the newborn baby: a reappraisal. Lancet 1985;2:137-40. Hosmer DW, Lemeshow S. Applied logistic regression. New York: John Wiley; 1989. p. 720e4. Senat MV, Minoui V, Multon O, Fernandez H, Frydman R, Ville Y. Effect of dexamethasone and betamethasone on fetal heart rate variability in preterm labour: a randomized study. BJOG 1998;105: 749-55. Multon O, Senat MV, Minoui S, Hue MV, Frydman R, Ville Y. Effect of antenatal betamethasone and dexamethasone on fetal heart rate variability in growth retarded fetuses. Fetal Diagn Ther 1997;12:170-7. Ozdemir H, Guvenal T, Cetin M, Kaya T, Cetin A. A placebocontrolled comparison of effects of repetitive doses of betamethasone and dexamethasone on lung maturation and lung, liver, and body weights of mouse pups. Pediatr Res 2003;53:98-103. Whitelaw A, Thoresen M. Antenatal steroids and the developing brain. Arch Dis Child Fetal Neonatal 2002;83:F154-7. Crowley P. Prophylactic corticosteroids for preterm delivery. In: The Cochrane library. Issue 4. Oxford: Update Software; 1999. Baud O, Foix-L’Helias L, Kaminski M, Audibert F, Jarreau PH, Papiernik E, et al. Antenatal glucocorticoid treatment and cystic periventricular leukomalacia in very premature infants. N Engl J Med 1999;341:1190-6. Canterino JC, Verma U, Visintainer PF, Elimian A, Klein SA, Tejani N. Antenatal steroids and neonatal periventricular leukomalacia. Obstet Gynecol 2001;97:135-9. O’Shea TM, Kothadia JM, Klinepeter KL, Goldstein DJ, Jackson BG, Weaver RG 3rd, et al. Randomized placebo-controlled trial of a 42-day tapering course of dexamethasone to reduce the duration of ventilator dependency in very low birthweight infants: outcome of study participants at 1-year adjusted age. Pediatrics 1999;104: 15-21. Murphy DJ. Effect of antenatal corticosteroids on postmortem brain weight of preterm babies. Early Hum Dev 2001;63:113-22. Modi N, Lewis H, Al-Naqeeb N, Ajayi-Obe M, Dore CJ, Rutheford M. The effects of repeated antenatal glucocorticoid therapy on the developing brain. Pediatr Res 2001;50:563-4.
224 27. Reist M, Marshall KA, Jenner P, Halliwell B. Toxic effects of sulphite in combination with peroxynitrite on neuronal cells. J Neurochem 1998;71:2431-8. 28. Haynes LE, Griffiths MR, Hyde RE, Barber DJ, Mitchell IJ. Dexamethasone induces limited apoptosis and extensive sublethal damage to specific subregions of the striatum and
Spinillo et al hippocampus: implications for mood disorders. Neuroscience 2001;104:57-69. 29. Murphy BP, Inder TE, Huppi PS, Warfield S, Zientara GP, Kikinis R, et al. Impaired cerebral cortex gray matter growth after treatment with dexamethasone for neonatal chronic lung disease. Pediatrics 2001;107:217-21.