Keywords
periviable PROM - serial amnioinfusion - pulmonary hypoplasia
Prelabor rupture of membranes (PROM) is the rupture of membranes before the onset
of labor. It is rare for PROM to occur before the onset of viability. Is estimated
to occur in approximately 0.37% of all pregnancies, and it presents a challenging
clinical dilemma due to high likelihood of poor neonatal outcome, as well as significant
risks to maternal health. PROM, before the onset of viability is frequently associated
with subclinical intra-amniotic infection.[1]
[2] Patients with PROM before the onset of viability face significant risks, including
development of chorioamnionitis, preterm labor, abruption, and intrauterine demise.
After counseling, patients may elect to continue the pregnancy, and in the absence
of clinical signs of infection, labor or vaginal bleeding, the mainstay of therapy
involves expectant management until the onset of viability. This conservative management
typically involves inpatient admission, latency antibiotics, and steroids for fetal
lung maturity. This strategy however, does not address the issue of the associated
sequela for the fetus, of which include severe pulmonary hypoplasia and compressive
limb deformities. The exact incidence of pulmonary hypoplasia is unknown and likely
underreported. Pulmonary hypoplasia is potentially lethal for the neonate. Factors
that affect the severity of pulmonary hypoplasia in the newborn include gestational
age at PROM, duration of latency, and the amount of amniotic fluid present around
the developing fetus.[3] More than 20 years ago, serial transabdominal intra-amniotic infusions with warmed
saline were proposed as an additive management approach to patients with PPROM (Preterm
Prelabor Rupture of Membranes). While the data from these trials have been encouraging,
lack of safety data and adequate control groups currently prevent this strategy from
becoming widely accepted.[4] The case we present is an example of the novel use of serial transabdominal amnioinfusions
using oxacillin infused normal saline, in a patient with PROM before the limit of
viability.
Case Report
Thirty-five year old gravida 3 para 1–0-1–1 was referred to our maternal-fetal medicine
office after diagnosis of early midtrimester oligohydramnios. Oligohydramnios was
incidentally found during a routine anatomical survey ultrasound examination at 18
weeks of gestation. She was evaluated for preterm prelabor rupture of membranes (pooling
of amniotic fluid in posterior vaginal fornix, arborization ferning testing and nitrazine testing); this initial evaluation was negative. Two subsequent assessments again
showed no evidence of rupture of amniotic membranes. The patient was offered, but
declined intra-amniotic dye infusion testing at that time. An ultrasound at 206/7 weeks of gestation revealed a single live fetus and an amniotic fluid index of 2.8
cm. There was ultrasound evidence of membrane separation, suggesting a possible membrane
rupture distal to the cervical opening (“high amniotic leak”). Fetal kidneys and bladder
were seen on ultrasound and appeared normal. Her pregnancy was complicated by a history
of gestational diabetes, history of loop electrical excision procedure, herpes simplex
type 2, and maternal obesity (prepregnancy body mass index [BMI] 34 kg/m2). An episode of domestic violence had also been reported. She had early genetic screening
with cell free fetal DNA that showed an appropriate fetal fraction, sex chromosomes
XY, and low risk for trisomy 13, 18, and 21. A second trimester maternal serum α-fetoprotein was 2.16 MoM (Multiples of the Median). The patient underwent prenatal
counseling with neonatology and maternal-fetal medicine. She expressed wishes for
full neonatal resuscitation at 22 weeks and stated an understanding of the prognosis.
After counseling she was offered an amnioinfusion and intra-amniotic dye instillation
testing to confirm suspected diagnosis of PPROM. The patient proceeded with amnioinfusion
and intra-amniotic dye installation at 216/7 weeks of gestation. Under ultrasound guidance a 22-gauge-needle was inserted into
the intra-amniotic cavity, and 15 mL of clear yellow fluid was removed for purposes
of genetic and infection testing. 40 mL of warmed 0.9% sodium chloride solution containing
1,000 mg/L oxacillin was infused at this time. 5 mL of indigo carmine was added for
the intra-amniotic dye instillation test. A total of 400 mL of warmed 0.9% sodium
chloride solution containing 1,000 mg/L oxacillin was infused. The patient then ambulated
for 30 minutes and tampon testing confirmed rupture of fetal membranes. She was admitted
to the hospital for latency antibiotics and serial intra-amniotic infusions. Acyclovir
prophylaxis was started after clinical exam confirmed no active herpes lesions were
seen. She completed a 7-day-course of latency antibiotics consisting of ampicillin
and azithromycin. Betamethasone was administered at 23 weeks of gestation.
At 226/7 weeks of gestation, she underwent a second amnioinfusion. At this time, 800 mL of
warmed 0.9% sodium chloride solution containing 1,000 mg/L oxacillin were infused
to obtain a normal maximum vertical pocket (MVP).
The plan was for serial intra-amniotic infusions until 26 weeks of gestation. However,
at 236/7 weeks, the patient denied continued leakage of fluid and MVP was normal. At this
time, it was suspected that the amnion had resealed and decision was made to cease
further intra-amniotic infusions unless symptoms of leakage of fluid returned or oligohydramnios
again developed. Amniotic fluid testing returned with no evidence of intra-amniotic
infection. Genetic testing with karyotype and microarray also confirmed no genetic
abnormalities in the fetus. At 27 weeks of gestation she developed gestational hypertension
without clinical or laboratory evidence for preeclampsia.
The patient was expectantly managed in the hospital. Fetal status remained reassuring
with appropriate fetal growth on serial ultrasounds. Additionally, the MVP remained
normal for the duration of her pregnancy. At 330/7 weeks, she was given a second course of betamethasone. An external cephalic version
was performed at 34 weeks due to breech presentation. She underwent a successful induction
of labor at 342/7 weeks. She was given penicillin in labor for group B streptococcus prophylaxis. She
delivered a vigorous male infant, with a birth weight of 2,160 g, Apgar's scores of
9 at 1 and 5 minutes of life. The placenta weighed 390 g and had evidence of moderate
acute inflammation (grade 3) at the chorion, but no evidence of inflammation elsewhere
on the amnion, or umbilical cord. The newborns initial white blood cell count was
11.7 × 1,000 μL. He was given ampicillin and gentamycin for 48 hours until negative
blood cultures per our neonatal intensive care protocol. The newborn stayed in the
hospital for 9 days and had no immediate complications.
Discussion
Premature rupture of membranes near the limit of viability occurs in less than 1%
of all pregnancies.[5] Identified risk factors for PPROM before the limit of viability include a history
of preterm delivery or PPROM in previous pregnancy, short cervical length, cerclage,
intra-amniotic procedures, antepartum vaginal bleeding, low body mass index, low socioeconomic
status, illicit drug use and tobacco use.[6]
[7]
[8]
[9] Most studies of PPROM before the limit of viability are retrospective and only include
cases of expectant management. With a paucity of evidence to guide clinical management,
this diagnosis creates a challenging clinical dilemma for clinicians and patients
alike.
The expected clinical course in a patient with rupture of membranes before viability
is variable and depends on the gestational age at rupture, the length of latency period,
and the presence or absence of oligohydramnios.[5] Data from an investigation by Falk et al showed that length of latency does not
appear to vary by gestational age at rupture. They showed the median latencies of
8 days (range: 1–161 days) with PROM < 20 weeks, 4.5 days (range: 2–106 days) with
PROM between 20–21 weeks, and 12.0 days (range: 1–112 days) with PROM between 21–23
weeks.[10] Patients, who choose expectant management must be counseled on the serious maternal
and perinatal risks. Maternal risks include development of chorioamnionitis (31.8%)
and subsequent development of endometritis (1%), abruption (9.3%), sepsis (1%), death
(1/619 cases).[11] There is a high rate of perinatal death with 31.6% stillbirth rate, and 29.7% neonatal
death rate. Survival for a live birth in the setting of PPROM before the onset of
viability is estimated at 44%.[12] Beyond high rates of death among the survivors, there is a significant risk for
pulmonary hypoplasia and fetal skeletal deformities.
After appropriate counseling, patients that desire conservative management and are
absent of clinical signs of infection, labor, or vaginal bleeding, undergo the mainstay
of therapy—expectant management until the onset of viability, at which point inpatient
admission, latency antibiotics, and steroids are typically recommended.[2] This strategy however, does not address the issue of associated oligo or anhydramnios,
or the subsequent fetal sequela of severe pulmonary hypoplasia and compressive limb
deformities. The exact incidence of pulmonary hypoplasia is unknown and likely underreported,
but some data suggest that it is found in 9 to 20% of newborns delivered after periviable
PROM prior to 26 weeks of gestation. Pulmonary hypoplasia is lethal in 50 to 100%
of cases.[3]
[13]
[14]
[15] Factors that affect the risk of pulmonary hypoplasia in the newborn include gestational
age at rupture of membranes, duration of latency, and the amount of remaining amniotic
fluid. Although, there are no reliable methods for diagnosing pulmonary hypoplasia
in utero, several groups demonstrated association between persistent severe oligohydramnios
and worse perinatal outcome.[12]
The novel use of serial transabdominal amnioinfusion to prolong latency and prevent
the development of pulmonary hypoplasia was first introduced over 20 years ago.[16] Since this paper, further studies into this approach have shown promise. The theoretical
benefit of amnioinfusion or the introduction of physiologic solution into the amniotic
cavity is proposed to be 3-fold. First, it provides a dilution of preexisting intra-amniotic
bacteria. Second, it washes out and dilutes inflammatory cells and mediators (prostaglandins,
leukotrienes, cytokines, interleukins among others). Lastly, it increases the intra-amniotic
fluid volume and intrauterine pressure. In theory, washing out or diluting the preexisting
intra-amniotic bacteria and inflammatory cells may be beneficial to prolong the latent
period and the presence of fluid may promote lung development and prevent positional
contractures.[17] Some additional secondary benefits have been proposed and include increasing the
ability to test fetal genetics, improving ultrasound imaging and decreasing the risk
of cord compression. Recently, a Cochrane review assessed the efficacy of this approach
with the data from two randomized control trials and concluded that the small number
of subjects in those studies precluded a definitive answer in regards to the efficacy
of the intervention.[4] A systematic review and meta-analysis suggested a better shortterm prognosis in
women with PPROM, who underwent serial amnioinfusion as seen in the reviewed observational
studies, but not in the randomized control trials (RCT). The intervention group had
significant latency prolongation and improved perinatal and neonatal survival and
experienced less pulmonary hypoplasia. These results intensify in the significantly
lower gestational ages at rupture of membranes in the intervention group, compared
with the control group in the observational studies.[17] These results from the meta-analysis of RCTs, demonstrated a trend toward benefit,
but the results were not statistically significant. This is possibly because of lack
of power in these studies due to the small number of participants. This review also
revealed a large variation in the interventions timing, continuation, and other interventions
performed (hospitalization, antepartum monitoring, antibiotic prophylaxis choice,
and steroid administration).[17]
Our case report describes a unique approach to intra-amniotic fluid infusion in a
patient with PPROM before the onset of viability. This case adds a novel approach
to amnioinfusion with the addition of oxacillin to the intra-amniotic fluid infusion.
This technique has been used in other intrauterine intervention procedures such as
meningomyelocele repair. Administration of antibiotic in intra-amniotic fluid infusion
was used as part of the standard protocol for the Management of Myelomeningocele Study
(MOMS) trial and most institutions performing these procedures are using this mixture
to replace amniotic fluid prior to completing closure of the hysterotomy.[18] The addition of oxacillin solution may have aided in the significant prolongation
of latency for our patient. This conclusion cannot be drawn definitively but should
be considered in further study into the potential benefits of serial amnioinfusion
are undertaken. While PPROM before the onset of viability remains a rare event, it
has serious morbidity for both mother and fetus. Evidence continues to suggest that
serial intra-amniotic fluid infusion is a permissible treatment option for well counseled
patients choosing to continue their pregnancy.
