Obstetrics in the time of Ebola: challenges and dilemmas in providing lifesaving care during a deadly epidemic.

The World Health Organization (WHO) has declared the current Ebola epidemic an international public health emergency. The epidemic is concentrated in West Africa, an area of the world with an already poor health infrastructure and with some of the worst health indicators globally. The WHO estimated Sierra Leone, one of the affected countries, to have the highest maternal mortality rate in the world, 1100 per 100 000 live births, equating to a lifetime risk of 1 in 21 women dying during childbirth. The provision of adequate maternity services in Sierra Leone is a challenge at the best of times; however, in the context of an Ebola epidemic there are unique challenges in both appropriate diagnosis and safe treatment. Whereas regular health services should continue to operate during an epidemic, it is unclear how best to provide safe obstetric interventions. In any humanitarian emergency it is estimated that approximately 15% of pregnant women will encounter a potentially life-threatening complication, it is these women which pose the greatest challenges in diagnosis and treatment during the epidemic. Ebola is highly infectious and is spread through contact with human bodily fluids, including breast milk and sweat. As women are often caretakers of the sick, and during pregnancy are more likely to have recently attended a health clinic, they are a high-risk group for exposure to, and transmission of, Ebola. The mortality rate of infected pregnant women is very high, with perinatal mortality believed to approach 100%. Clinical diagnosis of Ebola is difficult: the history and symptoms are often general, especially in the early stages, and overlap with many other differential conditions (for example malaria or typhoid). In obstetrics there is a large crossover between the presentation of women with pregnancy complications and the alert symptoms for a suspected Ebola case. Spontaneous miscarriage, bleeding (including vaginal bleeding), abdominal pain, chest pain, joint pain, vomiting, stillbirth/intrauterine fetal death (IUFD), and fever are routinely encountered in isolation or combination as part of obstetrical referral criteria. The same symptoms can also form part of the ‘case definition’ for an Ebola alert, in particular when combined with a history of contact with Ebola (including suspicion) or attendance at funerals. Differentiating between a woman who presents with an obstetric emergency, where prompt intervention is necessary and potentially lifesaving, and a woman who is suspected of having Ebola, where invasive procedures should be absolutely limited, remains a major challenge in this field. The current system used to identify suspected Ebola patients relies on verbal history and temperature alone. This takes place in a ‘triage’ area before entry to the unit is permitted. Because of the associated hazards of close contact with an infected individual, the stratification of risk takes place before health workers consider performing exposure-prone procedures or the patient mixes with other inpatients; however, this method is not robust. The patient history is often unclear, particularly for women who arrive in extreme circumstances and may be unconscious or bleeding profusely. Furthermore, the history is occasionally not reported correctly out of fear by patients or their relatives (for example, if they had been to a traditional birth attendant before coming, or if they have attempted to terminate the pregnancy). Finally, during the epidemic patients may withhold contact histories out of fear of stigmatisation, isolation, or refusal of treatment. Suspected Ebola patients should be isolated first, and then tested, taking high-level infection control precautions. Because of poor infrastructure and limited access to