Published in the September 2013 issue of Today’s Hospitalist
WHEN CALLED to take care of medically ill pregnant women, one of the most important things a hospitalist can do is remember that a fetus needs a mother.
“If mothers don’t perfuse, babies don’t perfuse. If mothers don’t oxygenate, babies don’t oxygenate,” explained Karen Rosene-Montella, MD, professor and vice chair of medicine and director of obstetric medicine at the Alpert Medical School of Brown University in Providence, R.I. “Fetal well-being is completely dependent on maternal well-being. There is not as great a dichotomy between these two entities as we are led to believe.”
Speaking to hospitalists during a session at this spring’s Society of Hospital Medicine conference in Maryland, Dr. Rosene-Montella laid out some dos and don’ts for hospitalists, as well as a philosophy of advocacy for women who need medical care when they are pregnant.
Take, for instance, how to care for women with blood clots. Because pulmonary emboli are the leading cause of maternal mortality, Dr. Rosene-Montella said it’s important to find them. Sometimes, you’ll need CT scans, MRIs and MRVs.
“In a lot of institutions, the radiologists are reluctant to do these scans,” due to fears about radiation, said co-presenter Matthew Sermer, MD, professor of obstetrics, gynnaecology and medicine at the University of Toronto. “It’s a constant fight because they are worried about the fetus. But if the mother has an embolism, that’s a problem” for the fetus too.
It’s the hospitalist’s job, noted Dr. Rosene-Montella, “to encourage people to go ahead and do the study.” Most patients who die of VTE do so “because of a failure to investigate, because of the mistaken belief that radiology testing is contraindicated in pregnancy.”
In fact, she explained, the U.S. National Council on Radiation Protection recommends a total dose of under 5 rads (0.05 Gy) during pregnancy as total exposure. Most commonly used diagnostic imaging will expose patients to “well under 1 rad.”
“It is very unlikely that patients will get to that level,” she said, “A CT angio is well under an acceptable level. Always, the right answer is to do the test when there is a concern for PE.”
Because pregnancy is a hypercoagulable state, physicians need to vigilantly pursue any suspicions. Making this task particularly difficult, however, is the fact that clinical signs can be unreliable. D-dimers, for example, can’t be used to rule out an event because they are normally elevated during pregnancy. “If you think someone has a DVT and you are not finding it,” Dr. Rosene-Montella said, “order an MRV. Many DVTs are higher in the pelvis and may be missed with conventional ultrasound techniques.”
Once a clot is found, initial treatment should be “exactly what you would use for a nonpregnant woman: either unfractionated heparin or low molecular weight heparin (LMWH), neither of which cross the placenta,” she said. Doses may be greater, however, because a pregnant woman has more blood volume and increased renal clearance.
It’s “an unanswered question” if doses should be calculated based on the woman’s pre-pregnant or pregnant weight. “You will have to follow anti-Xa levels closely and continue heparin throughout gestation and for six weeks postpartum,” Dr. Rosene-Montella said. “And if you need an IVC filter, you need one. Same indications.”
One major difference in pregnancy is that warfarin shouldn’t be used for this indication. That’s difficult because there are no data on the use of the newer anticoagulants, which do cross the placenta, so the standard is still subcutaneous LMWH.
The philosophy of treating ill pregnant women just as aggressively as nonpregnant women can help guide hospitalists when ordering drugs.
“There are very few medications that are truly harmful to the baby,” said Dr. Rosene-Montella. Those include category X drugs, which the FDA notes are contraindicated in pregnancy. Instead, for the vast majority of medications, there are little to no data in pregnant women.
The question hospitalists should ask is “are drugs indicated or not?” Dr. Rosene-Montella said. “If they are, try to use the safest possible one.” Generally, drugs to avoid include ACE inhibitors, angiotensin receptor blockers (ARBs), tetracyclines, warfarin, isotretinoin (Accutane) and fluoroquinolones.
Hospitalists also need to think about altered pharmacokinetics during pregnancy, including increased renal clearance and volume of distribution and altered absorption and protein binding. While these can affect dosing choices, they shouldn’t affect your decision to treat sick women.
“Much more harm is done when you withhold treatment in pregnant woman than providing indicated care,” Dr. Rosene-Montella said. “If you can medically treat a mother, you need to. And almost always, it is the right thing for the baby too.”
Added Dr. Sermer, “The message is that the baby doesn’t do well when the mother is failing. If you have somebody who is ill, not to give them medical treatment, I believe, is malpractice.”
This is not to say that caring for acutely ill pregnant women isn’t fraught with difficulty, let alone anxiety. One of the most dangerous conditions hospitalists get called for “preeclampsia “can be difficult to diagnose and hard to treat.
It can also result in terrible outcomes for women and their babies. As Dr. Sermer said, hospitalists should keep in mind that a new event in the third trimester “is usually preeclampsia unless proven otherwise.”
Part of the problem is that preeclampsia is misunderstood, misnamed and misjudged. “The traditional definition is high blood pressure plus proteinuria, but it is so much more,” Dr. Rosene-Montella said. “Patients can have preeclampsia in the absence of hypertension or proteinuria. It is related to endothelial dysfunction and vasospasm. It affects the cardiovascular system, the neurologic system, the liver, the kidneys, blood and the pulmonary system.”
In addition, she said, preeclampsia is not a precursor to eclampsia. “Eclampsia is preeclampsia with seizures, and patients might present initially with seizures.”
Unfortunately, the clinical manifestations of preeclampsia may not be obvious “until we are four steps down the road,” she added. Her advice is to always consider preeclampsia when seeing a pregnant woman complaining of multiple symptoms. For instance, she said, when a pregnant patient complains of a visual disturbance, belly pain and headache, “that’s preeclampsia.”
When evaluating a pregnant woman with suspected preeclampsia, it’s key “even if you have no obstetrician around “to get a good fetal assessment, Dr. Rosene-Montella pointed out. That includes a fetal nonstress test, an ultrasound for fetal growth and amniotic fluid volume assessment, and possible umbilical artery Dopplers. Decreased amniotic fluid volume is “a good marker of placental profusion.”
“Timing of delivery is very dependent on fetal assessment and severity of maternal illness, in consultation with the obstetrician,” she said. And treatment of preeclampsia and eclampsia is delivery, but delivering the placenta does not immediately make the preeclampsia go away. It can also occur postpartum, and “many serious cases go unrecognized” until late in their course because they happen postpartum.
One common mistake that Dr. Rosene-Montella sees hospitalists make is to give fluid to women with preeclampsia in whom they see decreased urinary output. But pulmonary edema is a complication that occurs in between 2% and 4% of preeclampsias. “I would encourage everybody to sit on your hands when you see decreased urine output in this population,” she said. “You are better off vasodilating someone than adding fluid.”
This is a “tough one,” Dr. Sermer admitted, because “a number of the deaths from preeclampsia come from pulmonary edema and not from renal failure.” In treating preeclampsia’s hypertension, meanwhile, use the same drugs that you would use in a nonpregnant woman having a hypertensive crisis: labetalol, hydralazine, nifedipine, postpartum ACE inhibitors. Nitroglycerin and nitroprusside can be used as well. The goal is to get the blood pressure below 150/100. All deaths noted in a U.K. register of such patients occurred when blood pressures exceeded 160.
Although peripartum cardiomyopathy is rarer than preeclampsia, it can occur in the last month of pregnancy or up to five months postpartum. It can also be difficult to differentiate between peripartum cardiomyopathy and decreased left ventricular function related to preeclampsia.
It’s helpful to remember that most LV dysfunction related to preeclampsia “gets better very, very quickly,” Dr. Rosene-Montella said, “and peripartum cardiomyopathy does not.”
With this heart disease, she said, the message again is the same: Treat the mother just like any other dilated cardiomyopathy. One exception is to use hydralazine for afterload reduction, and avoid ACE inhibitors and ARBs until after delivery. You may also want to add anticoagulation because of the high risk of thromboembolism.
Pregnant patients who need to be cardioverted because of acute arrhythmias should likewise be treated as aggressively as if they weren’t pregnant. But remember that the electrodes that may be attached to the mother for fetal monitoring are metal; make sure you remove them before delivering a shock to prevent arcing.
And when treating pregnant patients, Dr. Rosene-Montella said, it is particularly important to listen closely to what patients say. “This population is basically young and healthy and always looks less sick than they are,” she said.
One recent patient, for instance, had reported to the resident that her vision hadn’t changed. “When I went in and saw her, the patient couldn’t see,” Dr. Rosene-Montella said. “I said to her, ‘So, you have had a change in your vision.’ And she said, ‘No. It was the same yesterday.’ I learned a lesson about how carefully you need to ask patients questions. I am continuously humbled by how we should talk to patients.”
Deborah Gesensway is a freelance writer who covers U.S. health care from Toronto.
“Help, I need somebody!”
OF ALL THE THINGS that terrify hospitalists about caring for sick pregnant patients, topping the list is the feeling that you’re all alone making tough decisions for patients you know little about in matters you had next-to-no education in.
Karen Rosene-Montella, MD, director of obstetric medicine, professor and vice chair of medicine at Brown University in Providence, R.I., and senior vice president for women’s services for Lifespan, a health care system, wants to change that. She and her colleagues at Brown are offering a 24/7 telephone backup service that hospitals, physician groups and insurers can purchase to specifically help hospitalists treat sick pregnant women.
The service negotiates cost individually with each customer, said Dr. Rosene-Montella, depending on what doctors or hospitals want and need. Options range from continuing education or Webinars to a telephone backup service and consulting on how to structure obstetric medicine services. (For information, contact Dr. Rosene-Montella directly at Karen_ Rosene_Montella@Brown.edu.)
Based on their number of deliveries, hospitals can anticipate how many inpatients they can expect to have who are pregnant and need nonobstetric care. Given that about 30% of women who become pregnant have a chronic medical illness, there will be “plenty of medical complications,” she explained, and hospitalists “will be in the middle.”
When a pregnant patient walks into the emergency department with a terrible asthma exacerbation or bad hypertension, “the stakes are high,” Dr. Rosene-Montella notes. “And it’s not something you see all the time, so you are not comfortable.” That’s as true for obstetricians, she points out, as for hospitalists.
With the subscription service, she said, the hospitalist sends a page through an online physician scheduling program (www.amion.com) to an internist with specialized training in the care of pregnant patients. The two then have a telephone conversation. Her group is also developing a telemedicine component to the service.
Telephone consultation works because the hospitalist seeing the pregnant patient is already comfortable with CHF, thrombosis or asthma care. The hospitalist “just doesn’t know what the interplay of pregnancy is,” Dr. Rosene-Montella said. “We do a lot of translation between hospitalists and obstetricians.”