FDA Urges Calm After Stimulants Linked To Sudden Death

Agency awaits two additional observational studies after acknowledging that a study it funded has methodological flaws.

FDA is awaiting the results of two new trials evaluating the risks of stimulant medications after concluding that a recent study showing an association between products for attention-deficit/hyperactivity disorder and sudden death had significant limitations.

The case-controlled study, published in the American Journal of Psychiatry and funded by the FDA and the National Institute of Mental Health, compared two groups of children that died during the late 1980s and early 1990s to determine whether or not stimulant medication may have played a role in their deaths. But often-times, the medical history information was based on parents' memories - sometimes a decade later - which FDA noted was unreliable.

The recently published study "tells you there may be a risk, but it doesn't tell you what this risk may be," CDER Office of Medical Policy Director Robert Temple said during a June 15 teleconference to discuss the AJP study. "The studies we're working on ... will give people that information."

The two additional studies - one in children and one in adults - are being conducted in collaboration with the Agency for Healthcare Research and Quality. The observational studies use automated claims data and electronic health records to compare sudden cardiac events, myocardial infarctions and strokes in patients.

The observational studies, which follow cohorts of people who use stimulant medications, will be available in fall; pediatric results are expected first with adult results few months later. Those studies, along with the AJP study, stemmed from advisory committee recommendations in 2006 to examine a possible link.

Medications Vs. Car Crashes

"We don't think there was anything careless about the methodology," Temple said of the flawed study, adding that a case-controlled study made sense and the results are no one's fault. "It's the nature of the beast."

The case-controlled study compared the use of stimulant medications in 564 healthy children who died suddenly and 564 children who died as passengers in a motor vehicle accident. The use of stimulant medication was determined from parents, medical examiners and toxicology reports. (The two groups of children were compared because the children all died suddenly and the cause of death was not a known health problem.)

Out of 564 healthy children who died suddenly, 10 were reported to be taking a stimulant medication at the time of death, and out of 564 healthy children who died in a motor vehicle accident, two were reported to be taking a stimulant medication at the time of death. The results were statistically significant and the study authors concluded that there may be an association between the use of stimulant medications and sudden death in healthy children.

But FDA says the study's limitations are too great to conclude anything. "The statistical significance is well enough, but it depends on a relatively small number of events," said Temple. "If the number was 10 versus three it wouldn't be statistically significance."

The first limitation is that a child's use of a stimulant medication for ADHD was determined many years after each child's death. The deaths occurred between 1985 to 1996, but the data on medication usage were collected from March 1997 to January 2008. This time lag may have resulted in reporting errors.

Also, the differences in cause of death (sudden death versus death from a motor vehicle accident) could have influenced the family or caregiver's recall of information on stimulant medication use at the time of death.

Finally, the low frequency of stimulant use in both groups, as well as possible differences in the type of post-mortem inquiry, could have a profound biasing effect on the results.

"I don't think you can characterize the study as reassuring, but it doesn't make the case that there is a real risk here," Temple said.

As such, the agency's recommendations for healthcare professionals remain unchanged and include follow all the current prescribing information, such as taking a medical history for cardiovascular disease, performing a physical exam with special focus on the cardiovascular system (including examination for the signs of Marfan syndrome), and considering obtaining further tests such as a screening electrocardiogram and echocardiogram.

Products affected by the recommendations include: Focalin, Focalin XR (dexmethylphenidate HCl); Dexedrine, Dexedrine Spansules, Dextrostat (dextroamphetamine sulfate); Vyvanse (lisdexamfetamine dimesylate); Desoxyn (methamphetamine); Concerta, Daytrana, Metadate CD, Metadate ER, Methylin, Methylin ER, Ritalin, Ritalin-LA, Ritalin-SR (methylphenidate); Adderall, Adderall XR (mixed salts amphetamine); Cylert (pemoline) and generics.

-Lauren Smith ([email protected])