Why are NICU drugs largely untested?
Ninety percent of medications used on premature babies haven't been studied on infants for safety
Ninety percent of medications used on premature babies haven't been studied on infants for safety
Exactly 10 weeks before her due date, Shannon Frick woke up bleeding. She was scared, but, by now, she was used to being scared. Her entire pregnancy had been fraught with terror. Although Shannon initially conceived twins, one baby boy died in utero when she was 18 weeks pregnant. Survival was far from certain for his brother because Shannon鈥檚 body wasn鈥檛 making enough amniotic fluid. 鈥淲e had been told that if he made it to birth, there would be an 80 percent chance of severe neurological and pulmonary defects,鈥� recalls Shannon, who is now 34 and lives with her family in Holden, Massachusetts. 鈥淚t was devastating. We were grieving the loss of one baby and having to think about how far we would go to save this one.鈥�
Shannon rushed to the University of Massachusetts Memorial Hospital, where her bleeding was diagnosed as a partial placental abruption, a condition in which the placenta rips loose from the uterus, which can be fatal for both mother and child. Liam was delivered via emergency Caesarean section at 30 weeks gestation. He weighed just two pounds, 10 ounces, but he was breathing on his own. The baby was whisked away to the neonatal intensive care unit (NICU) before Shannon got a chance to hold him. Still, she and her husband, Michael, allowed themselves to hope.
But, within a few hours, Liam began struggling for air. Doctors rushed to put him on a ventilator, and then a high-frequency oscillator, a kind of souped-up ventilator that forces air into a patient's lungs by shaking them violently. The baby鈥檚 oxygen continued to drop. Shannon was still in her recovery room when a neonatologist came in to explain that Liam鈥檚 circulatory system wasn鈥檛 adapting well to life outside the womb; his blood pressure was spiking and his heart rate was erratic. Shannon is a nurse, so she was aware of the risks of this condition, known as persistent neonatal pulmonary hypertension. 鈥淭hat term scared the ever-living hell out of me,鈥� she says. 鈥淚 asked the doctor, 'Are you sure you can handle this?'鈥�
鈥淚 know I can,鈥� the neonatologist told her. But they would need to try treating Liam with nitric oxide, an inhaled medication that relaxes muscles in the lungs to improve respiratory function. The doctor explained that the Food and Drug Administration had approved nitric oxide for use in full-term babies—but it wasn鈥檛 approved for preemies like Liam. There just wasn鈥檛 enough data to go on. The FDA doesn鈥檛 regulate a doctor鈥檚 ability to prescribe drugs to an unapproved patient population, but going 鈥渙ff label鈥� would mean they had no idea of the potential side effects or complications. Then he added: 鈥淏ut if we don鈥檛 do it, Liam probably won鈥檛 live through the next 24 hours.鈥�
Shannon didn鈥檛 hesitate. 鈥淚 don鈥檛 care that it hasn鈥檛 been tested,鈥� she said. 鈥淚f my other option is walking out of this hospital without a baby, we need to do it.鈥�
Their gamble appeared to pay off. Liam began to stabilize over the next few hours, and he was able to wean off the ventilator four days later. But the next morning, an X-ray showed that the baby had developed a small brain bleed. Shannon and Michael knew it was likely a nitric oxide side effect. They looked at each other over Liam鈥檚 incubator, both thinking the same thing: Oh God. This is our fault.
The Fricks were working off little information, but they were lucky to know as much as they did about their doctor鈥檚 decision. 鈥淣eonatalogists and pediatricians have become so accustomed to prescribing a drug off-label, they hardly even think about it,鈥� says Robert Ward, M.D., emeritus professor and former director of the University of Utah School of Medicine鈥檚 pediatric pharmacology program. 鈥淎nd they usually don鈥檛 tell the parents what they鈥檙e doing.鈥�
That鈥檚 because 90 percent of the medications used in NICUs, as well as the majority used across all pediatric medicine, have never been tested for safety or efficacy in babies, especially those born before full gestation. If doctors stuck solely to on-label prescribing, they wouldn鈥檛 be able to treat most of the patients admitted to their care.
Liam recovered from the brain bleed. But Shannon and Michael couldn鈥檛 stop asking each other: Did we do the right thing? There was no way to know for sure. Liam spent 10 weeks in the hospital鈥檚 NICU. Now 3 years old, he is small for his age, continues to have respiratory issues, and is still catching up on speech and other milestones. 鈥淏ut he鈥檚 here,鈥� says Shannon today. Without the nitric oxide treatment, she鈥檚 sure he wouldn鈥檛 be. Yet she鈥檚 still shaken by their high-stakes decision. 鈥淲hy isn鈥檛 this medication studied so it can be approved for use in these babies?鈥� she says.
We're used to thinking that the United States offers the most cutting-edge healthcare in the world, yet countries like Lithuania and Bosnia have a lower than we do. While the reasons for that , a prime one is a higher rate of illness at birth. And our renowned medical experts are often working on their tiniest, most vulnerable patients without a net—relying on little more than raw intuition about which medicine to prescribe.
The FDA hasn鈥檛 approved a new drug specifically for neonatal use in 17 years. Christina Bucci-Rechtweg, M.D., head of pediatric and maternal health policy for Novartis Pharmaceuticals, is quick to acknowledge the 鈥渄esperate need鈥� for better research on pediatric drugs, but she鈥檚 also blunt about why it鈥檚 not happening: money. 鈥淭his makes me sad to talk about, but it鈥檚 very expensive to develop a therapy that鈥檚 only [for] a very small population,鈥� she explains. 鈥淵ou might not see a return on your investment, [and that鈥檚] been one of the main things keeping industry away from this field.鈥� In other words: You can sell erectile dysfunction drugs to tens of millions of adult men, but there鈥檚 no profit in medications that save the lives of a few hundred thousand preterm babies. 鈥淚t breaks my heart,鈥� says Shannon. 鈥淲e鈥檙e talking about tiny, little humans who have done nothing wrong but might not get a fighting chance because companies don鈥檛 support this research.鈥�
And because companies are reluctant to invest in new drug research, most of the drugs used in NICUs today are older and off patent, meaning any manufacturer can make them. Pharmaceutical companies profit off new drugs, which they can patent and sell exclusively for three to seven years, according to . So there鈥檚 no financial incentive to get better safety data on older drugs, when companies are already selling them as cheaply as possible.
But advocates for neonatal drug research argue that the pharmaceutical business model shouldn鈥檛 trump public-health math. When a healthy baby leaves the NICU, the cost of his or her treatment can then be amortized over a 70-year lifespan—while treating the heart condition of a 60-year-old will extend their life by maybe 10 or 15 years, tops. 鈥淚鈥檝e tried to make the case to both drug companies and the FDA that the lifetime impact should elevate the importance of this research,鈥� says Francis Sessions Cole, M.D., director of the division of newborn medicine at the Washington University School of Medicine in St. Louis, Missouri. But pharmaceutical companies aren鈥檛 as interested in 6-month-olds who are decades away from being a part of the larger drug-buying public. 鈥淭he argument doesn鈥檛 wash when there鈥檚 no money to be made,鈥� Cole notes. 鈥淎nd yet these babies are all unbelievably sick, and we need to do something.鈥�
When we don鈥檛, the worst can happen. In 2006, three premature babies died in a single week after they were accidentally given adult-sized doses of heparin, a blood thinner used to flush catheters and prevent blood clots, at Methodist Hospital in Indianapolis, Indiana. The FDA notes that the safety and efficacy of heparin in pediatric patients 鈥渉as not been established,鈥� and even includes a specific warning about the special risks it can pose when used in newborns. (News outlets that adult-sized doses of the drug had mistakenly been stored in the NICU鈥檚 drug cabinet; the hospital issued a statement offering to provide counseling and restitution to the families.)
These kinds of horror stories are why Rob and Alicia Furman of Raymore, Missouri, were anxious every time a medication arrived in the NICU for their daughter. Briella was born at just 23 weeks gestation on March 28, 2016, at Saint Luke鈥檚 Hospital in Kansas City, Missouri, weighing only 12 ounces. 鈥淲e were lucky to be at the only hospital in our area that stocks small enough breathing tubes,鈥� says Rob. 鈥淚f we鈥檇 been somewhere else, they might not have even tried to resuscitate her.鈥� Nearly every one of the 156 days Briella spent in the NICU brought a new fight to keep her alive. Every time the Furmans asked Briella鈥檚 doctors about the potential risks of a treatment, they say they received the same unsettling answer: 鈥淭here鈥檚 just no research on that.鈥�
The powerful opiate fentanyl was one under-researched drug prescribed to Briella for pain management. Fentanyl is not FDA-approved for use in patients under age 2 because there are no studies establishing its safety and efficacy, and the Furmans were alarmed to see the hospital pharmacy send the drug over in an adult-sized, 50-cc syringe. "The dose for Briella was only half a cc," says Alicia. "Our medical team was incredible, but every time they gave it to her, I thought about how it would be so, so easy to mis-dose these tiny babies."
That鈥檚 assuming doctors know the right dose to give in the first place. Every morning, Malgorzata Michalowska-Suterska attends rounds in intensive care units of Maria Fareri Children鈥檚 Hospital at Westchester Medical Center in Valhalla, New York, where she works as a clinical pharmacy specialist. Residents balance laptops and coffee mugs; attending physicians study X-rays and fire off questions. Michalowska-Suterska follows with an iPad, making meticulous notes about every change in a patient's medication list.
Often a doctor will pause to ask her which type of diuretic or mood stabilizer Michalowska-Suterska thinks would be the best match for that particular child. All she can do is make her best guess.
鈥淲hen my colleagues come over from the adult side of the hospital, they can鈥檛 believe this is how we do things,鈥� she says. 鈥淲hen an adult is put on a new medication, we know how well it will work because we have these wonderful randomized, placebo-controlled, double-blind studies done on 2,000 patients.鈥� She鈥檚 referring to the 鈥済old standard鈥� research trials required by the Food and Drug Administration for any new drug to hit the market: Large rosters of patients are randomly assigned to groups that receive either the medication in question or a placebo as a control. The trials are known as double-blind, because neither patients nor doctors will know which pill they鈥檙e taking, which lets researchers analyze their results without bias. If patients taking the drug fare better than patients taking a placebo (while meeting other study targets, like a low rate of side effects), the drug is likely to be approved for use in patients like those in the study.
Meanwhile, in the pediatric wards: 鈥淲e have case reports with two patients,鈥� says Michalowska-Suterska. 鈥淥r very small studies. If you can find a trial with 15 patients, you think, Wow, this is great.鈥� When a doctor asks her to recommend a drug, she looks at research done on adults, and then estimates how to titrate a dose for a one- or two-pound preemie.
But estimates only get you so far. About one third of infants experience what doctors and pharmacists euphemistically call a pharmokinetic surprise, where they develop side effects not usually seen in adult patients—or the drug simply doesn鈥檛 work because the dose wasn鈥檛 strong enough, says Matthew Laughon, M.D., a neonatologist and professor of pediatrics at the University of North Carolina at Chapel Hill. This 鈥渟urprise鈥� rate is so high because babies are not tiny adults: Every system in their body is still developing, and they circulate and metabolize drugs in unpredictable ways. Whether you鈥檙e 20, 40, or 60, an adult's kidneys clear drugs out of the body in more or less the same way—and much more effectively than the still-developing kidneys of a newborn. 鈥淲hen you鈥檙e talking about babies born 10 or 14 weeks early, maturation after birth is unpredictable,鈥� says Laughon.
The Furmans say they understand why most of the medications given to baby Briella had no relevant research: 鈥淭he odds of survival when you鈥檙e born at 23 weeks are just 25 percent,鈥� says Rob. 鈥淓ven if you could find four babies like her to study, three of them aren鈥檛 going to make it. There goes your data.鈥� Every year, only 500,000 infants are born prematurely in the United States. When you start sorting them by medical diagnoses, the lists get even shorter.
In addition to slicing into drug company profits, that makes it hard to find a truly representative sample of babies with the same type of congenital heart defect, or the same type of postnatal respiratory issue. And many parents are understandably skeptical about the premise. Ward recalls trying to study methadone as a painkiller in newborns; he knew from preliminary data that it has fewer side effects than many other pain medications. But over three years, he was only able to recruit seven children to participate in his study. 鈥淧arents would literally back away from me,鈥� he says. 鈥淭hey would say, 鈥楾hat will make my child an addict.鈥欌�
Researchers also have to decide how to control for a long list of variables: How was the mother鈥檚 prenatal health care? What other drugs has the infant received? And they have to decide on a reasonable end point to the research—a difficult task when the goal is to save a baby鈥檚 life and give them a relatively normal lifespan. 鈥淲e鈥檙e always asking, how long is long enough?鈥� says Jonathan Davis, M.D., chief of newborn medicine at the Floating Hospital for Children at Tufts Medical Center in Boston. Depending on the drug, researchers may need to follow their trial subjects for years to show that a medication given at birth doesn鈥檛 impact how they learn to walk, talk, or do long division. And the longer a study runs, the more expensive it becomes.
Of course, physicians and pharmacists can draw on their own clinical experience for these decisions. Shannon Frick says she felt reassured when Liam鈥檚 neonatologist revealed he had successfully used nitric oxide in a baby with similar problems just the week before. But a doctor鈥檚 anecdotal experience always involves a smaller sample size and a higher degree of confirmation bias than a well-run trial. And when researchers are finally able to look closely at frequently used drugs, they often discover problems. Laughon points to cisapride, a medication used to treat gastric reflux by moving stomach contents along the digestive tract, which was removed from the market after decades of use when a study finally revealed that it could cause sudden death in babies. 鈥淸We've had] multiple situations where we鈥檝e gone back to study drugs after we鈥檝e used them and found out that, lo and behold, they鈥檙e really not safe,鈥� says Davis. 鈥淎nd it鈥檚 [a] shame on us that we didn鈥檛 really study it in the first place.鈥�
Davis has been fighting for those studies for most of his career. He now chairs the Neonatal Advisory Committee for the FDA鈥檚 Office of Pediatric Therapeutics, implementing what he calls 鈥渁 system of carrots and sticks鈥� created as a result of several laws passed over the past 15 years and made permanent under the . The 鈥渃arrot鈥� dangled before pharmaceutical companies is a six-month extension on the exclusivity of any new drug patent if they study the drug in pediatric patients. 鈥淎n extra six months on patent can translate to millions of dollars in exclusive sales,鈥� says Ward. 鈥淭hat can more than offset the cost of the research.鈥�
The 鈥渟tick鈥� is that the FDA can now require drug makers to study a medication in children any time they submit an adult application. And they have to consider testing their new drugs specifically in newborns (not just older kids, who are easier and cheaper to study). 鈥淚f you鈥檙e not studying the drug in a neonatal population, you have to be able to say why on your application,鈥� says Susan McCune, M.D., director of the FDA鈥檚 Office of Pediatric Therapeutics.
As a direct result of this legislation, 675 medications have had their labels updated to reflect new research on the correct dosing guidelines for pediatric patients. But only a fraction of those drugs will help treat newborns, counters Ward. The rest are geared toward older infants and children. He and many others argue that the current regulations don鈥檛 go far enough. 鈥淭o some extent, the law has prompted drug companies to focus on developing drugs that don鈥檛 have any use in children,鈥� says Cole. 鈥淭hat lets them avoid the issue all together.鈥� Bucci-Rechtweg of Novartis doesn鈥檛 disagree: 鈥淭he diseases of neonates pretty much don鈥檛 exist in adults,鈥� she says. 鈥淰ery few new adult therapies are being developed that may [work for] the neonatal population.鈥�
But while the pharmaceutical industry parses their legislative mandates for loopholes, babies in every NICU and PICU in the country are being injected with under-researched medications. After Rachel Thomas鈥檚 water broke while 16 weeks pregnant with her son Zachary, she spent the next 10 weeks on bed rest, trying to stave off labor. 鈥淚 was advised to have an abortion, because the odds of his survival were so grim,鈥� she says. 鈥淲e were in a world of hell.鈥�
Zachary was delivered via C-section at Inova Fairfax Hospital in Falls Church, Virginia, 14 weeks before his due date, and immediately taken away for intubation. Like Shannon, Rachel hadn鈥檛 even held Zachary when she was asked to consent to the use of nitric oxide within hours of his birth. Zachary stabilized on the ventilator but wasn鈥檛 able to breathe on his own; within a few days, his kidneys began to shut down. A doctor took Rachel and Chris aside and explained that there were drugs they could try. 鈥淏ut you should know your rights as parents,鈥� he added. 鈥淚f at any point you don鈥檛 want us to continue, it鈥檚 your right to tell us to stop and just give you time to hold him.鈥�
It was the moment that Rachel had spent her whole pregnancy trying not to imagine but knew would probably come—when she and Chris had to decide whether to keep fighting to save their baby or let him die in their arms. She looked at Chris and then at the doctor. 鈥淲hatever drug you need to give him, just do it,鈥� she said. "We're going to throw the kitchen sink at this.鈥� During the seven months he spent in the hospital鈥檚 NICU, Zachary was prescribed 40 different drugs.
Three years later, Rachel can鈥檛 remember which combination the doctors tried to treat her baby鈥檚 kidney failure—but she does remember feeling numb with relief when Zachary finally wet his diaper, a sign that his kidney function was returning. It was one of many days when she left his incubator to go to the NICU鈥檚 private breast pumping room and cry. 鈥淲hen the decisions are that black or white, you don鈥檛 think about side effects or long-term risks,鈥� she says now. 鈥淎nd it鈥檚 hard to explain the fog we were moving through in those first few weeks; I was trying to stay in this positive state of denial to make it through my pregnancy, and now that was all crashing down.鈥� She was recovering from a C-section, pumping breast milk around the clock, and trying to figure out how to be a mother to a baby she could barely touch. Now, when Rachel wades through the long list of Zachary鈥檚 prescriptions, she says, 鈥淢ost of these, I never even knew he was given.鈥�
That may be the crux of the matter. Parents thrown into catastrophic infant health crises need to feel informed and able to trust their medical teams to make evidence-based decisions. But they鈥檙e also operating through a haze of fear, exhaustion, uncertainty and grief—and they don鈥檛 know what they don鈥檛 know. 鈥淚 don鈥檛 [tell] parents upfront that there are major gaps in our understanding of drug metabolism in babies,鈥� says Cole, noting that he doesn鈥檛 usually share that information until a baby experiences an adverse effect. 鈥淭hen we retroactively explain that we did this based on the best information we had.鈥�
The right information will require years of new research and millions of research dollars. In the meantime, hundreds of thousands of kids like Liam, Briella, and Zachary are growing up as survivors—the tiniest of babies who overcame incredible obstacles. 鈥淏ut beating these kinds of odds comes with a price,鈥� says Rachel. Four years later, Zachary is a sweet, curious boy. But he鈥檚 just beginning to speak in complete sentences. He also struggles with low muscle tone and significant feeding and motor-skill delays. Rachel often wonders which of her son鈥檚 challenges are related to his early birth, and which could be the lingering consequences of all of those drugs that saved his life. 鈥淚 mean, something caused these things,鈥� she says. 鈥淏ut I鈥檒l just never know.鈥�
