Association of histological chorioamnionitis and magnesium sulfate treatment in singleton and dichorionic twin pregnancies with preterm premature rupture of membranes: preliminary observations

Introduction

Inflammation of the chorioamniotic membranes (chorioamnionitis) is an etiologic factor for preterm premature rupture of the chorioamniotic membranes (pPROM), premature uterine contractions, prematurity, and perinatal mortality and morbidity including cerebral palsy [1].

In the USA, 800,000 children receive active treatment for disorders of movement and posture with or without cognitive impairment [2]. The lifetime cost to care for an individual with these conditions recognized under the term of cerebral palsy (CP) is nearly $1 million, with a combined lifetime costs for all individuals with CP born in the year 2000 estimated at $11.5 billion [3], [4], [5]. Prevention of preterm birth and neuronal cell damage caused by cerebral ischemia, cerebral hemorrhage or excessive inflammatory response in premature newborns using oxygen radical scavengers, calcium channel antagonists and mild hypothermia is clinically and socioeconomically important [6], [7].

Antenatal short-term treatment with magnesium sulfate (MgSO₄) in patients at risk for delivery prior to 34 weeks’ gestation has also been recommended to reduce the risk for CP [8], [9], [10], [11]. However, the five randomized control trials (RCTs) that addressed the effectiveness of antenatal exposure to MgSO₄ for neuroprotection had different conclusions [12], [13], [14], [15], [16]. Furthermore, these RCTs provided little information concerning patient selection, optimal timing for treatment or dosage of MgSO₄ in patients admitted to the hospital with pPROM.

MgSO₄ is an electrolyte replenisher that inhibits the N-methyl-D-aspartate (NMDA) receptor and stimulates nitric oxide synthase activation with vasodilation, immunomodulation, neurotransmission and muscle excitatory actions [7], [11], [17], [18], [19]. In pregnancy, the MgSO₄ vasodilator side effects manifest with some degree of maternal diaphoresis, flushing, nausea, vomiting, palpitations and headache with risk for neuro-muscular blockage, respiratory depression and respiratory or cardiac arrest at maternal MgSO₄ blood concentrations above 10 mEq/L (5.0 mmol/L) [7], [20]. MgSO₄ toxicity is recognized in neonates with respiratory depression, hypotonia, and hyporeflexia [21], [22], [23], [24].

Elimian and collaborators were the first to report the association between antenatal treatment with MgSO₄ and clinical chorioamnionitis in a study of 401 patients with singleton pregnancies and pPROM between 23 and 34 weeks’ gestation (190 cases exposed to MgSO₄ and 211 not exposed to MgSO₄; OR=2.8, 95% CI=1.14, 6.90; P=0.02) [25]. Kamyar and collaborators in 2016 reported a multivariable analysis of data from 367 children that had antenatal exposure to MgSO₄ and concluded that the neuroprotective effects of antenatal MgSO₄ treatment was lost when clinical chorioamnionitis was present [26]. Our aim in this study was to gain further insights into the possible association of antenatal MgSO₄ treatment with histological chorioamnionitis in two groups of patients with pPROM: (1) singleton and (2) dichorionic twin pregnancies with one intact amniotic sac.

Materials and methods

This was an observational cohort study using the medical records of patients delivered before 37 weeks’ gestation between January 2008 and January 2012 at Jersey Shore University Medical Center, a small tertiary care medical center in the State of New Jersey. For the purpose of the study we selected those cases admitted to the hospital with pPROM prior to 34 weeks’ gestation. Exclusion criteria included admission after 34 weeks’ gestation, delivery before 24 weeks’ gestation, intrauterine fetal demise and monochorionic twins. All cases were identified using the hospital ICD-9 coding database. The study was approved by the Jersey Shore University Medical Center institutional review board.

Management of patients with pPROM was in accordance with the recommendations provided in the ACOG practice bulletin published in 2007 [27]. At our center, antenatal treatment with MgSO₄ was recommended for seizure prophylaxis in pre-eclamptic patients and for neuroprotection prior to 34 weeks’ gestation [6], [9], [10], [11]. Patients treated with MgSO₄ are perceived to be at risk for delivery. The authors recognize that today many institutions use 32 weeks’ gestation or less as the cutoff for MgSO₄ treatment for neonatal neuroprotection. During the study period, we followed published recommendations not exceeding a 4-g loading dose, 1–2 g/h maintenance dose, and 24 h for the duration of treatment. The safety checklist for MgSO₄ treatment before anticipated preterm birth for neuroprotection was completed at the time of the patient’s admission to the hospital [28]. MgSO₄ treatment for neuroprotection was similar for patients with singletons and twins.

Sociodemographic and clinical variables assessed from the patients’ medical records included age, parity, gravidity, gestational age at admission, latency period of pPROM in days, treatment with MgSO₄, antibiotic treatment, steroids treatment, vaginal birth, primary cesarean delivery, repeat cesarean delivery, birthweight, Apgar scores at 1 and 5 min, and composite perinatal complications defined by the presence of an intrapartum non-reassuring fetal heart rate tracing or severe neonatal complications including IVH or any type of organ failure. Antibiotic treatment was not included as a confounder because patients received at least 1 dose prior to delivery.

Histological chorioamnionitis was the primary study outcome. Chorioamnionitis can be categorized as subclinical, clinical and histological. Subclinical chorioamnionitis is detected by direct analysis of the amniotic fluid IL-6, glucose concentration, WBCs, Gram stain, cultures, and molecular genetics methods including proteomics. Clinical chorioamnionitis is defined by maternal fever together with maternal leukocytosis >15,000; tachycardia >100 bpm, uterine tenderness, foul odor of amniotic fluid, or fetal tachycardia >160 bpm. Despite clinical chorioamnionitis being treated with antibiotics about half of patients still demonstrate intra-amniotic bacterial colonization. Histological chorioamnionitis requires presence of clusters of inflammatory cells such as neutrophils, eosinophils or lymphocytes in the gestational sac membranes or chorionic plate in H&E stain (ICD-10: O41.1; ICD-9: 658.4, 762.7) [29], [30].

Results

Among 107 patients that had preterm deliveries during the study period, 57 admitted to the hospital before 34 weeks’ gestation with pPROM were included in the analysis. Fifty cases were excluded because they were admitted after 34 weeks’ gestation, ended with intrauterine fetal demise, delivered before 24 weeks’ gestation or had monochorionic twins.

The mean maternal age was 30 years, the median gravity 2 and the median parity 1 (Table 1). Twenty-four (42.1%) of the 57 patients with pPROM before 34 weeks’ gestation were treated with MgSO₄. The majority of the 57 patients (80.1%) delivered before 34 weeks’ gestation and none of the 11 patients that delivered after 34 weeks’ gestation received MgSO₄ treatment. Overall 28% of patients had a vaginal delivery, 54% had a primary cesarean delivery and 18% a repeat cesarean delivery.

Thirty-six patients had singleton pregnancies (63%) and 21 had dichorionic twin pregnancies (37%). In the singleton pregnancy group 24% had vaginal deliveries, 57% primary cesarean deliveries and 15% repeat cesarean deliveries. In the twin group, 30% had vaginal deliveries, 39% primary cesarean deliveries and 24% repeat cesarean deliveries (Table 2). Histological chorioamnionitis was reported in 15 of the singleton cases (41%) and in 4 of the twin cases (19%). In twins, histological chorioamnionitis was reported only in the placenta with the ruptured amniotic sac. Composite perinatal complications occurred in 24% of patients with singleton pregnancies and 43% of patients with twins.

The logistic regression model identified antenatal exposure to MgSO₄ as a significant predictor for histological chorioamnionitis in singleton pregnancies after controlling for the clinical confounders (χ²=6.46; P=0.01). Such association was not observed among women with dichorionic twin pregnancies with one intact amniotic sac. No statistical differences were noted between singletons and twins for latency period, gestational age at delivery, birthweight, Apgar score <7, or for severe neonatal morbidity.

Discussion

We found a statistically significant association between histological chorioamnionitis and antenatal MgSO₄ treatment prior to 34 weeks’ gestation in patients with singleton pregnancies and pPROM after controlling for socio-demographic and clinical confounders. This observation is aligned with a previous report that concluded that there was a higher rate of clinical chorioamnionitis in patients that received antenatal MgSO₄ treatment [25]. Our novel and surprising finding, however, was a dissimilarity in the rates of histological chorioamnionitis between singleton and dichorionic twins with one intact amniotic sac (41% in the singleton group and 19% in the twin group). This dissimilarity could not be attributed to (1) differences in the recommended management protocol for MgSO₄ neuroprotection, (2) length of the latency periods between the 2 groups, or (3) maternal age, gravidity, parity, gestational age at delivery, or birthweights (Tables 1 and 2). Our observations therefore, pave the way for future investigations aimed to identify modulators of MgSO₄ biological actions in pregnancy.

Prematurity, defined as birth before 37 weeks’ gestation, affects 11.4% of all deliveries in the USA with an estimated $26.2 billion annual societal economic cost in the year 2005 [30], [31]. Preterm PROM is the leading cause of prematurity [1]. Antenatal MgSO₄ treatment is recommended for seizure prophylaxis in pre-eclamptic patients and to reduce the risk for cerebral palsy in children born before 34 weeks’ gestation. Indeed, in 1995, Nelson and collaborators reported a case-control study of 3-year-old children born at 30 weeks’ gestation in which moderate to severe CP was less likely when the child had been exposed antenatally to MgSO₄ (odds ratio of 0.14 of CP (95% CI, 0.05–0.51) [32]. Three observational studies that followed were not as conclusive [33], [34], [35]. Five randomized control trials (RCTs) addressing the effectiveness of antenatal exposure to MgSO₄ for neuroprotection were also indecisive [12], [13], [14], [15], [16].

In the MagNet study [12] conducted in 2002, 149 patients seen between 24 and 34 weeks’ gestation in preterm labor were randomized into four research arms, two tocolytic arms (with and without MgSO₄) and two non-tocolytic arms receiving either a single bolus of 4 g of MgSO₄ with no maintenance infusion or saline infusion. Although there were no cases of CP in the tocolytic MgSO₄ group, the authors concluded that the numbers were too small to detect a statistical difference in patients treated with MgSO₄.

The Australian Collaborative Trial of MgSO₄ Collaborative Group, analyzed 1062 surviving children delivered before 30 weeks’ gestation after randomization into receiving MgSO₄ or a 0.9% saline solution. At a corrected age of 2 years, the authors found that the combination of death/substantial gross motor dysfunction (17.0% vs. 22.7%; relative risk 0.75; 95% CI, 0.59–0.96) was statistically reduced in the MgSO₄ group. Importantly, they also found that MgSO₄ had no serious harmful effects [13].

The Magpie trial aimed at the long-term effects of MgSO₄ for 3283 children whose mothers had severe pre-eclampsia before 37 weeks’ gestation [14]. Mothers were randomly assigned to receive either MgSO₄ or placebo. The follow-up consisted of a questionnaire, with interview and neurodevelopmental testing at 18 months of age. Of those exposed to MgSO₄, 15.0% were dead or had neurosensory disability compared with 14.1% in those receiving placebo (relative risk 1.06, 95% CI 0.90–1.25). There were no substantial differences in causes of death or in the risk of individual impairments or disabilities.

The PREMAG trial, included 688 infants born at less than 33 weeks’ gestation in 18 French tertiary hospitals [15]. Mothers were randomly assigned to either MgSO₄ or 0.9% saline solution. Infants exposed to MgSO₄ had a reduced mortality rate of 9.4% vs. 10.4% (OR: 0.79, 95% CI 0.44–1.44, NS); severe white-matter injury 10.0% vs. 11.7% (OR: 0.78, 95% CI, 0.47–1.31, NS), and combined adverse outcomes 16.5% vs. 17.9% (OR: 0.86, 95% CI, 0.55–1.34, NS) relative to controls. A significantly protective effect of MgSO₄ was only found for combined severe motor dysfunction or death (OR: 0.62, 95% CI, 0.41–0.93).

The National Institute of Neurological Disorders and Stroke, National Institute of Child Health and Human Development, and the Maternal Fetal Medicine Units Network performed the largest RCT to date, titled the “Beneficial Effects of Antenatal Magnesium Sulfate (BEAM)” trial. In this study, 2241 patients between 24 and 31 weeks’ gestation age at risk of imminent delivery were randomized to MgSO₄ treatment or placebo. Follow-up was completed for 95.6% of their children. The composite of perinatal death, death by 1 year of corrected age or moderate CP at or beyond 2 years of corrected age, was not significantly different in the MgSO₄ group compared to the placebo group (11.3% and 11.7%, respectively; RR: 0.97; 95% CI, 0.77–1.23). In a secondary analysis, moderate or severe CP was statistically less frequent in the MgSO₄ group compared to the placebo group (1.9% vs. 3.5%; RR: 0.55; 95% CI, 0.32–0.95) [16].

Three meta-analyses that included these RCT trials were then published [8], [9], [36] and based on these reports, the consensus in perinatal medicine is that short-term antenatal exposures to MgSO₄ has neuroprotective actions; the number of women that would need to be treated with MgSO₄ to prevent one case of CP is 63 [9], [11].

Elimian and collaborators were the first to report the association of clinical chorioamnionitis with MgSO₄ in patients with pPROM without a clinical effect on any of the neonatal morbidity or mortality outcome measures [25]. Kamyar and collaborators then reported a multivariable analysis of data from 367 children and concluded that antenatal MgSO₄ treatment was not associated with a decrease CP at 2 years of age in cases with clinical chorioamnionitis [26]. These observations together with our findings support the hypothesis that prior to 34 weeks’ gestation, the rate of chorioamnionitis could increase in some patients with antenatal MgSO₄ treatment and that this effect could mitigate the neuro-protective benefits of MgSO₄ treatment in the neonate.