Optimized supportive care

In 2018, a multidisciplinary panel of international experts formulated evidence-based recommendations for delivering optimized supportive care to individuals with acute EVD. Recommendations were formulated based on the quality of evidence as well as benefits, harms, values, and preferences. The majority of recommendations, including rehydration therapy, systematic vital sign assessment, and biochemistry assessment were strongly recommended to all patients with suspected, probable or confirmed EVD as steps to reduce mortality and optimize care (7). These recommendations were adapted for the WHO guidance document “Optimized supportive care for Ebola virus disease” for adults and children (8).

Investigational therapies

Three published reports described investigational therapeutic use in pregnant women with acute EVD (3,9,10). During the 2014-2016 outbreak, van Griensven et al. conducted a non-randomized comparative study using convalescent plasma. Among 84 participants, eight pregnant women were treated with convalescent plasma in addition to supportive care. Mortality was 25% among pregnant women and 32% among non-pregnant individuals after receiving plasma treatment. Compared to historical controls with EVD, one of the two (50%) pregnant women died from EVD, but pregnancy in the control group was incompletely documented (3).

A pregnant woman with EVD was also treated with favipiravir in addition to supportive care during the 2014–2016 (9,10). The woman went into preterm labour and died from haemorrhagic shock. Her newborn child tested positive for EBOV and was treated with investigational therapies. The child became the first known survivor of congenitally-acquired EVD, and an examination at 12 months of age revealed normal development and a complete recovery from EVD (9,10).

The safety and efficacy of certain investigational therapeutics (ZMapp, Remdesivir, mAb114, REGN-EB3) are being evaluated in an ongoing randomized study. Pregnancy was not used as exclusion criteria in this study (11). Interim analyses noted improved outcomes resulting from treatments including REGN-EB3 or mAb114, and because of this, early termination of the trial was recommended. The efficacy of these treatments in pregnant women specifically is not yet known (12).

Additionally, the WHO convened an expert panel regarding investigational therapeutics using the MEURI ethical framework. The panel recommended that among vulnerable populations such as pregnant women, investigational therapeutics should be considered, and that until evidence suggests otherwise, pregnant women should be offered similar treatments to the non-pregnant population. Given that very few infants born to women with EVD survive, teratogenic effects were considered secondary to the mother’s health; however, risks and benefits should be highlighted in informed consent procedures (6).

Disease transmission

Among studies describing the role of IPC measures at time of childbirth for EVD-positive women, multiple lapses in IPC measures likely contributed to the transmission of EBOV from pregnant women to others. These included inadequate and improper use of PPE such as using contaminated material for multiple patients, point of care exposure from patients with unrecognized EVD, and lack of proper isolation procedures (17–19). Contacts who subsequently contracted EVD after lapses in IPC measures included healthcare workers and caregivers (18), and other patients at health-care facilities (19).

In Sierra Leone, investigators performed contact tracing of a pregnant woman with EVD after she presented in labour. She delivered via caesarean at a general hospital and developed EVD symptoms two days following delivery. She died shortly after and a post-mortem buccal swab confirmed the diagnosis of EVD. The infant developed symptoms at three days of life and died seven days later. Investigators identified 46 individuals who had contact with the pregnant woman during her stay at the health-care facilities, and among these individuals, six contracted laboratory-confirmed EVD. In this study, investigators revealed that the nurses, hospital cleaning staff and caregivers did not have access to recommended PPE. Other potential contributors to the high risk of exposure from this case included a lack of recognition of EVD infection, exposure from infected bodily fluids, and lack of isolation precautions (18).

Similarly, Connolly and Young performed contact tracing on a pregnant woman with undiagnosed EBOV who delivered in a maternity ward in 2014. Investigators identified multiple lapses in IPC measures that likely contributed to EVD infection for two maternity ward patients and their infants. One of the maternity patients who contracted EVD was delivered by the same nurse as the index patient; PPE used by the nurse was both incomplete and contaminated from the index patient. Further, the maternity patient’s delivery suite was the same as the index case and was not cleaned between deliveries. The second maternity patient was likely infected because of a lack of standard precaution procedures, as she was placed next to the index case during her delivery and subsequent haemorrhage (20).

Detectable EBOV RNA has been identified in amniotic fluid (21,22), placental tissue (21,23,24), foetal tissue (17, 21, 25, 26,27), vaginal secretions (28), and even in pregnant women with mild or unrecognized disease (17,27).

Viral persistence in pregnancy-related fluids and tissues

Persistence of EBOV has been documented in pregnancy-related fluids and tissues after clearance of the virus from blood has occurred (17,21) (Table 1). EBOV was successfully isolated from a foetal tissue sample in a woman who delivered a stillborn foetus one month after being exposed to EVD (17). Similarly, a woman from Sierra Leone delivered a stillborn foetus who tested positive for EBOV RNA, yet denied EVD or exposure to it. Her blood was EBOV RNA negative, IgM antibody negative, and IgG antibody positive for EBOV (27).

In reproductive tract specimens from non-pregnant individuals, EBOV RNA was detected up to 36 days following symptom onset for EVD (29), but all samples from menstrual blood in non-pregnant women tested negative (30).

Ebola in breastmilk

EBOV RNA has been detected in the breastmilk of women with acute and convalescent EVD up to 26 days after symptom onset (31,32), as well as in women with asymptomatic EVD (33,34) (Table 1). Two lactating women were evaluated after their infants died from laboratory-confirmed EVD. EBOV RNA was detected in their breastmilk by RT-PCR despite negative testing in blood (33,34).

Of 25 infants who were breastfed by mothers with EVD, 68% developed presumed or laboratory-confirmed EVD, with a mortality rate of 82%. Eight infants were breastfed by EVD-positive women but did not become ill (31–38). Other studies of four breastfed infants did not identify if the mother or infant developed EVD first (31,33,34). However, Bower et al (35) found that of 14 mothers who clearly acquired EVD prior to their breastfed infants, 86% (n=12) of the exposed infants contracted EVD, with EVD status proving to be the greatest risk. Breastfeeding alone was not identified as a risk factor for EVD transmission.

Table 1

Viral persistence in pregnancy-related fluids/tissues and breastmilk, of Ebola virus RNA (detected by RT-PCR) and if viral isolation was attempted and successful.