American Journal of Obstetrics and Gynecology (2005) 192, 4338

www.ajog.org

GENERAL OBSTETRICS AND GYNECOLOGY: OBSTETRICS

The combined maternal administration of magnesium

sulfate and aminophylline reduces intraventricular
hemorrhage in very preterm neonates
Gian Carlo Di Renzo, MD, PhD,a,* Marcella Mignosa, MD,a Sandro Gerli, MD,a
Liliana Burnelli, MD,a Giuseppe Luzi, MD,a Graziano Clerici, MD,a
Fabiana Taddei, MD,a Doretta Marinelli, MD,b Patrizia Bragetti, MD,b
Daniele Mezzetti, MD,b Benedetta Della Torre, MD,b Alessandra Fantauzzi, MD,b
Maria Serena Lungarotti, MD, PhDb
Centre of Perinatal and Reproductive Medicinea and the Neonatal Intensive Care Unit,b University of Perugia,
Perugia, Italy
Received for publication April 22, 2004; revised July 29, 2004; accepted July 30, 2004

KEY WORDS
Intraventricular
hemorrhage
Preterm labor
Aminophylline
Corticosteroids
Magnesium sulfate
Preterm premature
rupture of
membranes

Objective: To determine whether the adjunctive administration of aminophylline and magnesium

sulfate to mothers at risk for preterm birth can reduce the rate of intraventricular hemorrhage in
neonates born at less than 30 weeks of gestation.
Study design: A prospective study was conducted to determine whether the rate of
intraventricular hemorrhage was different in patients at risk for preterm delivery treated with
ritodrine, magnesium sulfate, aminophylline, and corticosteroids (group A) versus patients
treated with ritodrine and corticosteroids (group B). During the study period (January 1996 to
December 2001), 125 patients enrolled in the study. Treatment was assigned by alternative
allocation, and the study was designed to compare the rate of intraventricular hemorrhage in
neonates born before the 30th week of gestation (primary outcome), 78 newborns in group A and
68 in group B. The proportion of neonates with intraventricular hemorrhage was calculated, and
data were analyzed with Student t test, c2, and logistic regression analysis.
Results: The frequency of severe respiratory distress syndrome needing surfactant replacement
and high-pressure positive ventilation, patent ductus arteriosus, and retinopathy of prematurity
was not different between the 2 groups. However, the rate of intraventricular hemorrhage was
lower in neonates born before 30 weeks whose mothers received adjunctive aminophylline and
magnesium sulphate (group A) than in the group that did not receive these 2 agents (group B).
The overall frequency of intraventricular hemorrhage was 5.1% (4/78) versus 20.6% (14/68)
(P ! .001), and the frequency of intraventricular hemorrhage grade 3-4 was 1.3% (1/78) versus
10.3 % (7/68; P! .001), respectively.

* Reprint requests: Gian Carlo Di Renzo, MD, PhD, Center of Perinatal and Reproductive Medicine, Department of Gynecologic, Obstetric and
Pediatric Sciences, University of Perugia, Policlinico Monteluce, 06122 Perugia, Italy.
E-mail: direnzo@unipg.it
0002-9378/$ - see front matter 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.ajog.2004.07.078

434

Di Renzo et al
Conclusion: Adjunctive maternal administration of aminophylline and magnesium sulfate was
associated with a significant reduction in the rate of intraventricular hemorrhage in neonates born
before 30 completed weeks.
2005 Elsevier Inc. All rights reserved.

The likelihood of survival of very preterm neonates

has improved dramatically over the last 20 years.1
However, long-term neurologic handicap continues to
be a very substantial problem.2,3 Several strategies have
been proposed to decrease the risk of brain injury
associated with preterm birth and neonatal intensive
care. For example, corticosteroids administration reduces the rate of intraventricular hemorrhage (IVH),4
whereas acute treatment with magnesium sulfate appears to reduce the frequency of neurologic injury that
predisposes to cerebral palsy.5 This hypothesis was
tested after uncontrolled trials suggesting that antenatal
magnesium sulfate administration may be protective
against cerebral palsy,6 although other studies7-10 found
magnesium to be only weakly or not at all associated
with decreased risk of cerebral palsy.
Animal experimentation indicates that the administration of a phosphodiesterase inhibitor, aminophylline,
may improve lung adaptation at birth and that the
operative mechanism may be similar to that of badrenergic agents or glucocorticoids.11,12 Moreover,
aminophylline has a stimulatory eect on the central
nervous system, including the units involved in the
regulation of breathing.13
This study was conducted to explore whether maternal administration of aminophylline and magnesium
sulfate could reduce the rate of IVH in very early
preterm neonates.

Material and methods

A prospective study was conducted to determine
whether the administration of aminophylline and magnesium sulfate reduces the rate of IVH in preterm
neonates born at less than 30 weeks of gestation. The
study period was January 1, 1996, to December 31,
2001. The study was approved by the Institutional
Review Board of the Centre of Perinatal and Reproductive Medicine of the University Hospital in Perugia
(Project 27/95) with appropriate informed consent.
Patients at risk for preterm delivery admitted to the
Centre of Perinatal and Reproductive Medicine were
considered for inclusion into the study. Only patients
with a gestational age of less than 30 weeks were eligible
to participate. Dating of pregnancy was based on
a midtrimester ultrasound, which was routinely performed between 20 and 22 weeks. Pregnant patients at
risk for preterm delivery because of spontaneous preterm delivery, spontaneous preterm premature rupture

of membranes, and complications of pregnancy that

could lead to an indicated preterm delivery were invited
to participate. Patients admitted to the hospital during
the study period were informed of the study and invited
to participate. Informed consent was obtained. Allocation of the patients to either treatment was accomplished by alternative allocation to either group A or B.
This was not done by random allocation.
Ritodrine was administered at a rate of 0.10-0.30 mg/
minute by increments of 0.05 mg/minute every 10
minutes above the initial infusion rate as recommended
by the manufacturer or until contractions ceased. In the
case of preterm premature rupture of membranes,
ritodrine infusion was used at 0.10 mg/minute for at
least 48 hours, even in patients without evidence of
increased uterine contractility. At the same time that
ritodrine was administered, another intravenous line
was established to administer 4 g of magnesium sulfate
and 240 mg of aminophylline at a rate that would have
maintained the infusion for 12 hours (5.5 and 0.33 mg/
minute, respectively), continuing the same infusion
every 12 hours, for a minimum of 48 hours or until
delivery. Corticosteroids were administered as recommended by the National Institutes of Health Consensus
Conference (betamethasone 12 mg intramuscularly 24
hours apart).14 This group was considered group A.
Patients allocated to group B received ritodrine and
corticosteroids only, according to the same regimen
described in the previous paragraph. Patients with
preterm premature rupture of membranes received antibiotics administration (ceftriaxone 1 g intramuscularly
every 12 hours) until delivery or for a maximum of
a week. The study was designed to evaluate the
frequency of IVH, and our pre hoc hypothesis was that
this benet would accrue in neonates born below 30
weeks of gestation that had received a full course of
steroids. Therefore, infants born after 30 weeks or
before the completion of the course of steroids were
excluded. If a neonate was diagnosed to have a congenital anomaly not diagnosed prenatally or if there was
a fetal death, the case was also excluded from analysis.
Analysis was performed using Student t test, c2 test, and
multiple logistic analyses as appropriate.

Results
During the study period, there were 78 newborns in
group A and 68 in group B who met the inclusion
criteria. Maternal and neonatal characteristics and route

Di Renzo et al
Table I

435

Perinatal characteristics of study population

Maternal characteristics*
Multiple pregnancy (twin-triplets)
PIH-Pre-eclampsia-HELLP
Severe IUGR (!3 centiles)
PPROM
Route of delivery
Vaginal delivery
Cesarean section
Mean Apgar
1 min (range)
5 min (range)
Birth weight (g)
Range
Mean G SD
Timing of delivery (from admission) and length of therapy
Range (d)
Mean G SD
Mean gestational age at delivery
Range (wks)
Mean G SD
Male/female ratio

Group A (65 pregnancies,

78 newborns)

Group B (60 pregnancies,

68 newborns)

13
14
10
32

12
10
9
27

27 (34.6%)
51 (65.4%)

21 (35.3%)
44 (64.7%)

NS

1-9
6-10

1-9
5-10

NS

565-1220
757 G 215

600-1260
821 G 275

NS

2-18
7.2 G 5.4

2-16
8.1 G 4.8

NS

23-30
26.8 G 2.7
34/44

24-30
27.2 G 2.5
31/37

NS

Significance

NS

NS

NS, Nonsignificant; PIH, pregnancy-induced hypertension; HELLP, hemolysis, elevated liver enzymes, low platelets; IUGR, intrauterine growth restriction;
PPROM, preterm premature rupture of membranes.
* More than one may have been present in the same pregnancy.

of delivery are summarized in Table I. All newborns

were inborn and immediately transferred to the neonatal
intensive care unit after stabilization in the delivery
room.
The incidence of respiratory distress syndrome
(RDS) (severe enough to require surfactant replacement
and high-positive pressure ventilation),15 IVH (total
and grade 3-4),16 patent ductus arteriosus (PDA), and
retinopathy of prematurity (ROP) in the 2 groups is
reported in Table II. A highly signicant decrease in the
incidence of IVH was observed in group A (either total
number or 3-4 degree), compared with group B. The
frequency of neonatal death within 28 days from
delivery was not signicantly dierent between the 2
groups.
Possible confounding factors of the observed decreased association between treatment A and IVH was
evaluated by multiple logistic regression, entering variables singly and in combination. The following variables
were included into the models: gestational age (as
a continuous variable or dichotomized); birth weight
(continuous or dichotomized to those !750 g versus
750-1000 g versus O1000 g), small for gestational age
(SGA; dened as those born with a birth weight less
than 3 centiles for that gestational age) versus non-SGA,
year of birth, gender, clinically diagnosed chorioamnionitis, mode of delivery (cesarean section versus vaginal),

presence of labor, and neonatal surfactant treatment.

None of these factors signicantly changed the association between the rate of IVH and treatment with
magnesium sulfate and aminophylline.

Comment
Principal findings of our study
The key observation of this study is that the combined
used of aminophylline and magnesium sulphate was
associated with a reduction in the rate of IVH (total
frequency as well as the rate of grade 3 and 4) in very
preterm infants (those born before the 30th week of
gestation). Adjunctive treatment with this combination
did not change the rate of other complications such as
RDS, PDA, and ROP. We think that our observation
justies further studies of the potential value of magnesium sulfate and aminophylline in the prevention of
neurologic injury.

IVH and its clinical significance

Preterm neonates are at increased risk for IVH, and the
risk is higher the lower the gestational age at birth.
Hemorrhage often originates in the subependymal
germinal matrix, but it often ruptures through the

436

Di Renzo et al

Table II

Neonatal mortality and morbidity according to different antenatal treatments

IVH (3-4 degree)

Total
RDS*
PDA
ROP
Neonatal deathy

Group A (78 newborns)

Group B (68 newborns)

Significance

1
4
28
7
2
8

7
14
26
5
4
7

P ! .001
P ! .001
NS
NS
NS
NS

(1.3%)
(5.1%)
(35.9%)
(9.0%)
(2.6%)
(10.2%)

(10.3%)
(20.6%)
(38.2%)
(7.5%)
(5.9%)
(10.3%)

* Severe degree needing surfactant replacement and high-pressure positive ventilation (HPPV).
y
Within 28 days from delivery.

ependyma into the ventricular cavity.17 The germinal

matrix undergoes dramatic changes during the late
midtrimester including proliferation of glial cells and
oligodendrocytes. These changes are thought to require
a rich vascular supply. The germinal matrix disappears
as term approaches, and it is believed that the vascular
network that supplies it is not bestowed with an
adequate extracellular matrix envelope and smooth
muscle that would confer stability to the blood vessels.18
Other factors implicated in the etiology are a delay in
endothelial cell tight junction formation and pathophysiologic conditions such as uctuations in cerebral blood
ow (from hypercarbia and hypotension from systemic
neonatal hypotension).19 The bleeding can be aggravated by the suboptimal hemostasis (platelets and
coagulations factors) reported in some preterm neonates.19 Bleeding may lead to periventricular infarction
(white matter damage), which has been associated with
the subsequent development of cerebral palsy.17-19 No
eective preventive and/or therapeutic strategies have
yet been developed other than the administration of
steroids.18

Magnesium and aminophylline decrease IVH in

preterm neonates born before the 30th week
of gestation
The combined use of magnesium and aminophylline was
associated with a decrease frequency of IVH in very
preterm babies (less than 30 weeks of gestational age)
but not of severe RDS, PDA, ROP, or neonatal death.
The potential eect of several confounding factors was
evaluated with logistic regression, but none was found
to change the results. This study represent the rst
clinical trial in humans and is considered as preliminary
evidence because it was an observational, unblinded
study in which assignment was made by alternative
allocation. We have no evidence that alternative allocation introduced biases, but we considered that a randomized controlled trial (RCT) is optimal and the next
step in testing this hypothesis. Before the conduction of
the study, we planned to restrict the analysis to neonates
born before the 30th week of gestation because after that

gestational age, there is a dramatic reduction in the

frequency of IVH. The recent RCT of magnesium
sulfate for prevention of cerebral palsy was also targeted
to neonates born before the 30th week.5

Magnesium sulphate as a neuroprotective agent

Magnesium sulfate is widely used in obstetrics for
seizure prophylaxis in pre-eclampsia20 or for tocolysis.21
Evidence from randomized clinical trials supports its use
in pre-eclampsia, whereas the eectiveness of this agent
for tocolysis is less strong. Recently there has been
considerable interest in the utilization of magnesium
sulphate as a neuroprotective agent. Although the
evidence from nonrandomized clinical trials is contradictory,22,23 the only RCT reported to date suggests that
magnesium sulphate administration to the mother for 24
hours before birth reduced signicantly substantial gross
motor dysfunction (6.6% in the control group and 3.4%
in the treatment group; relative risk, 0.51, 95% condence interval, 0.29-0.91) in babies born before the 30th
week of gestation.5 Mechanisms by which magnesium
sulfate might be neuroprotective include stabilization of
both myocardium and blood ow in placental and fetal
cerebral circulation24 and a reduction in the area of
ischemia-reperfusion injury.25 Benets might be also
derived from blockage of the excitatory neurotransmitter receptors, an antioxidant eect of magnesium and
a reduction in platelet adhesiveness.10,26

Antepartum use of aminophylline as

a neuroprotective agent
Antenatal administration of aminophylline has been
used for the treatment of acute asthma,27 tocolysis,28
and the treatment of fetal distress associated with
uterine hyperdynamic state.29 Xanthines are widely used
in neonatology for the prevention and/or treatment of
neonatal apnea30 because they increase the respiratory
drive.31 Aminophylline crosses the placenta readily,
and serum theophylline concentrations in maternal
venous and mixed cord blood at delivery are almost
identical after intravenous bolus administration of
aminophylline.32 There is also no evidence of neonatal

Di Renzo et al
toxicity when aminophylline is administered in the third
trimester of pregnancy.33 Thus, maternal aminophylline
administration may be a method to deliver this agent to
the fetus. Previous studies reported by our group34
documented that short-term maternal administration
of aminophylline to pregnant rabbits at 26 days improved the survival rate and the respiratory performance of the pups, which were delivered by hysterotomy
far from term (mean gestational duration in rabbits is 31
days). These observations were interpreted as indicating
that aminophylline administration induces maturation,
which is similar to that induced by steroids. Because the
mechanisms of action of steroids and aminophylline are
dierent, we proposed that a synergistic eect could occur
and hence the use of aminophylline in this study. Other
investigators have also demonstrated that short
treatment with aminophylline of preterm neonates is
associated with a signicant increase in oxyhemoglobin.35
In conclusion, we present the rst evidence that the
combined administration of magnesium sulfate and
aminophylline can reduce the rate of IVH in very
preterm neonates. This benet was observed in neonates
who had received steroids (both arms of the study), and
therefore the combination of magnesium sulfate and
aminophylline appears to oer additional advantages
over the current standard approach, which is limited to
steroid administration. The pharmacologic approach
reported herein is relatively simple and inexpensive,
and the doses utilized have a large margin of maternal
safety (8 g per day of magnesium sulfate and 480 mg/day
of aminophylline). Further studies are required to
dissect the precise contribution of magnesium sulfate
or aminophylline to improve neonatal outcome as well
as to conrm our ndings. This can be accomplished
only through a large multi-institutional, randomized
clinical trial.

References
1. Evidence report/technology assessment no. 18: management of
preterm labor. AHRQ publication no. 01-E021, October 2000.
Available from: www.ahrq.gov.clinic/epix.htm.
2. Clark SL, Hankins GDV. Temporal and demographic trends in
cerebral palsydfact and ction. Am J Obstet Gynecol 2003;188:
628-33.
3. Winter S, Autry A, Boyle C, Yeargin-Allsopp M. Trends in the
prevalence of cerebral palsy in a population-based study. Pediatrics 2002;110:1220-5.
4. Crowley P. Prophylactic corticosteroids for preterm birth. Cochrane Database. Syst Rev 2000:CD000065.
5. Crowther CA, Hiller J, Doyle L, Haslam R. Eect of magnesium
sulfate given for neuroprotection before preterm birth. A randomized controlled trial. JAMA 2003;20:2269-76.
6. Leviton A, Paneth N, Susser M, Reuss ML, Allred EN, Kuban K,
et al. Maternal receipt of magnesium sulfate does not seem to reduce
the risk of neonatal white matter damage. Pediatrics 1997;99:1-5.
7. Nelson KB, Grether JK. Can magnesium sulfate reduce the risk of
cerebral palsy in very low birthweight infants? Pediatrics 1995;
95:263-9.

437
8. Canterino JC, Verma UL, Visintainer PF, Figueroa R, Klein SA,
Tejani NA. Maternal magnesium sulfate and the development of
neonatal periventricular leucomalacia and intraventricular hemorrhage. Obstet Gynecol 1999;93:396-402.
9. Mittendorf R, Dambrosia J, Dammann O, Pryde PG, Lee KS,
Ben-Ami TE, et al. Association between maternal serum ionized
levels at delivery and neonatal intraventricular hemorrhage.
J Pediatr 2002;140:540-6.
10. Schendel DE, Berg CJ, Yeargin-Allsopp M, Boyle CA, Decoue P.
Prenatal magnesium sulfate exposure and the risk for cerebral
palsy or mental retardation among very low-birth-weight children
aged to 3 to 5 years. JAMA 1996;276:1805-10.
11. Karotkin EH, Kido MK, Cashore WJ, Redding RA, Douglas WJ,
Stern L, et al. Acceleration of fetal lung maturation by aminophylline in pregnant rabbits. Pediatr Res 1976;10:722-4.
12. Barrett CT, Sevanian A, Phelps DL, Gilden C, Kaplan SA. Eects
of cortisol and aminophylline upon survival pulmonary mechanics,
and secreted phosphatidyl choline in prematurely delivered rabbits.
Pediatr Res 1978;12:38-42.
13. Cosmi EV. Drugs and anesthetics in the management of preterm
labor and delivery. In: Cosmi EV, editor. Obstetrical anesthesia
and perinatology. New York: Appleton-Century-Crofts, 1981:
547-623.
14. NIH Consensus Development Panel on the Eect of Corticosteroids for Fetal Maturation on Perinatal Outcomes. Eect of
corticosteroids for fetal lung maturation on perinatal outcome.
JAMA 1995;273:413-8.
15. Stark AR, Frantz ID. Respiratory distress syndrome. Pediatr Clin
North Am 1986;33:533-44.
16. Volpe JJ. Intracranial hemorrhage of the premature infant.
Neurology Newborn 1987;11:331.
17. Berger R, Garnier Y. Pathophisiology of perinatal brain damage.
Brain Res Rev 1999;32:107-34.
18. Berger R, Garnier Y, Jensen A. Perinatal brain damage: underlying mechanisms and neuroprotective strategies. J Soc Gynecol
Investig 2002;9:319-28.
19. Whitelaw A. Intraventricular haemorrhage and post haemorrhagic
hydrocephalus: pathogenesis, prevention and future directions.
Semin Neonat 2001;6:135-46.
20. Magpie Trial Collaboration Group. Do women with preeclampsia
and their babies benet from magnesium sulfate? The Magpie
trial: randomized placebo controlled trial. Lancet 2002;359:
1877-80.
21. Grether JK, Hoogstrate J, Walsh-Greene E, Nelson KB. Magnesium sulfate for tocolysis and risk of spastic cerebral palsy in
premature children born to women without preeclampsia. Am J
Obstet Gynecol 2000;183:717-25.
22. Mittendorf R, Covert R, Boman J, Boman J, Khoshnood B,
Lee KS, et al. Is tocolytic magnesium sulphate associated with
increased total paediatric mortality? Lancet 1997;350:1517-8.
23. Mittendorf R, Pryde P, Khoshnood B, Lee KS. If tocolytic
magnesium sulphate is associated with excess total pediatric
mortality, what is the impact? Obstet Gynecol 1998;92:308-11.
24. Mittendorf R, Pryde PG. An overview of the possible relationship
between antenatal pharmacologic magnesium and cerebral palsy.
J Perinat Med 2000;28:286-93.
25. Garnier Y, Middelanis J, Jensen A, Berger R. Neuroprotective
eects of magnesium on metabolic disturbances in fetal hippocampal slices after oxygen-glucose deprivation. J Soc Gynecol
Investig 2002;9:86-92.
26. Elimian A, Verma R, Ogburn P, Wiencek V, Spitzer A, Quirk JG.
Magnesium sulfate and neonatal outcomes of preterm neonates.
J Matern Fetal Neonatal Med 2002;12:118-21.
27. Adams BK, Cydulka RK. Asthma evaluation and treatment.
Emerg Med Clin North Am 2003;7:587-91.
28. Liu DTY, Blackwell RJ. The value of a scoring system in
predicting outcome to preterm labour and comparing the ecacy

438
of treatment with aminophylline and salbutamol. Br J Obstet
Gynaecol 1978;85:418-24.
29. Baxi LV, Petrie RH. Pharmacologic eects on labor: eects on
drugs on dystocia, labor and uterine activity. Clin Obstet Gynecol
1987;30:19-32.
30. Miller D, Schmidt B. Controversies surrounding xanthine therapy.
Semin Neonatol 2004;9:239-44.
31. Gerhardt T, McCarthy J, Bancalari E. Eects of aminophylline on
respiratory center and reex activity in premature infants with
apnea. Pediatr Res 1983;17:188-91.

Di Renzo et al
32. Romero R, Kadar N, Gonzales Govea F, Hobbins JC. Pharmacokinetics of intravenous theophylline in pregnant patients at term.
Am J Perinatol 1983;1:31-5.
33. Ramzan I, DeDonato V. Theophylline neurotoxicity in pregnant
rats. Life Sci 1988;42:491-5.
34. Cosmi EV, Saitto C, Barbati A, Del Bolgia F, Di Renzo GC,
Grossman G, et al. Eect of aminophylline on lung maturation in
preterm rabbit fetuses. Am J Obstet Gynecol 1986;154:436-9.
35. Lipshitz J. Uterine and cardiovascular eects of aminophylline.
Am J Obstet Gynecol 1978;131:716.