Swine flu and the prenatal clinic

A risk factor for swine flu complications is pregnancy. Yet one of the few venues specifically for healthy people in modern health care facilities is for pregnant and postpartum women, the in and out patient portions of the obstetrics department. And what of the newborn whose mother gets the flu? Infants are also at greater risk from influenza complications.

This is a tough problem. CDC has just issued some interim guidance regarding swine flu and obstetrics settings. The guidance is "interim" because CDC is quite frank that there is a great deal we don't know about the probabilities and severity of this virus in pregnant women. Still, some attempt to think this through is needed because the virus doesn't care how much we know and don't know. Here's a sketch of what they have come up with.

The underlying principle is to keep healthy people separated from those likely to be or possibly infectious. This means being able to identify sick patients, visitors or staff from those that aren't. If a pregnant woman hasn't been inclose contact with a known case of swine flu at home or at work, then they can be treated without special consideration. Those with identified exposure taking antivirals (Tamiflu or Relenza) for prophylaxis can be seen along with others but with standard infection control (presumably droplet precautions). But if they have an influenza-like illness (ILI; abrupt onset of fever greater than 100 degrees F., cough or sore throat, and no known other cause for the symptoms) they should be treated as if they have flu, even before testing. At this point they should be isolated from healthy women. This obviously means they should not be seen in the prenatal clinic until they are well (this isn't part in the guidance, but it is implied). If it is time to deliver their baby and the mother is still sick, CDC recommends the following:

Place a surgical mask on the ill mother during labor and delivery, if tolerable, in order to decrease exposure of the newborn, healthcare personnel, and other labor and delivery patients to potentially infectious respiratory secretions.

Place the ill mother in isolation after delivery (http://www.cdc.gov/h1n1flu/guidelines_infection_control.htm). The mother who has influenza-like-illness (http://www.cdc.gov/h1n1flu/casedef.htm) at delivery should consider avoiding close contact with her infant until the following conditions have been met: she has received antiviral medications for 48 hours, her fever has fully resolved, and she can control coughs and secretions. Meeting these conditions may reduce, but not eliminate, the risk of transmitting influenza to the baby. Before these conditions are met, the newborn should be cared for in a separate room by another person who is well, and the mother should be encouraged and assisted to express her milk. Breast milk is not thought to be a potential source of influenza virus infections. As soon as all conditions are met, the mother should be encouraged to wear a facemask, change to a clean gown or clothing, adhere to strict hand hygiene and cough etiquette when in contact with her infant, and begin breastfeeding (or if not able to breastfeed, bottle feeding). She should continue these protective measures, both in the hospital setting and at home, for at least 7 days after the onset of influenza symptoms (http://www.cdc.gov/h1n1flu/guidance_homecare.htm#c). If symptoms last more than 7 days, she should discuss the symptoms with her doctor. Protective measures might need to be continued until she is symptom-free for 24 hours. People who are once again well 7 days after getting sick are thought to be at low risk for transmitting the virus to others. (CDC Guidance: Considerations Regarding Novel H1N1 Flu Virus in Obstetric Settings, July 6, 2009)

If the mother has a flu infection, what about the newborn? We don't know, but CDC believes it is prudent to consider the baby potentially infected (and presumably infectious) for 2 days before onset of illness in the mother to 7 days after the start of illness. In those cases, CDC recommends appropriate infection control used for the newborn for its stay in the hospital (http://www.cdc.gov/h1n1flu/guidelines_infection_control.htm). The baby should be carefully watched for signs of flu and if suspected, immediate testing and initiation of antiviral treatment are recommended. Under an Emergency Use Authorization, oseltamivir (Tamiflu) can be used to treat these infants (but not for prophylaxis). While breat feeding during Tamiflu use by the mother is not contraindicated, if the baby is isolated from the mother as recommended it will have to be fed by hospital personnel using the mother's expressed breast milk. Visitors should be limited, taking into account "the patients’ emotional well-being and care."

As seen in this last item, there's still a lot of room for judgment and subjectivity in these guidelines (I tend to think this is good, rather than a weakness). They do what they are supposed to do: provide guidance. This is just one of the many difficult problems posed by epidemic infectious disease in the community. Contagious diseases propagate via social relations, but social relations are part of everything we do. The underlying principle is applicable to many other questions that arise: minimize, to the extent practicable and feasible, mixing people who are sick with people who are healthy, with special attention to those at most risk. This doesn't mean absolute quarantine of the exposed or hermetic isolation of those that are ill. When you tamper with established social relations you are in danger of creating even bigger problems than the ones you are trying to solve and you might well prevent measures that could have great net benefit.

It's a good general principle, but cannot be a governing one. We see in this instance one way to use it.