Child Psychiatrists Updated on Medication Side Effects

Sigita Plioplys, M.D., sometimes has her doubts about the Food and Drug Administration's (FDA) judgment.

In January 2008, the FDA issued an alert about increased risk of suicidality associated with all antiepileptic drugs, regardless of type.

“They said that all antiepileptics treat seizures, so they all must have the same side effects,” said Plioplys, head of the neuropsychiatric clinic at Children's Memorial Hospital in Chicago and an associate professor at the Feinberg School of Medicine at Northwestern University.

Classifying antiepileptic drugs is not so simple, Plioplys told listeners at the American Academy of Child and Adolescent Psychiatry's annual meeting in Honolulu last October. “They are different medications linked by a common indication, but then are differentiated by different mechanisms of action.”

Different drugs can thus produce the same results, but not necessarily the same side effects, she said.

The FDA review of 199 studies covering almost 44,000 patients found that four patients on the drugs committed suicide, compared with none on placebo.

“However, for the suicidality analysis, the FDA only included the one-third of the studies with positive suicidal events and excluded the rest,” she said. “These findings were not seen for all drugs, and there was no information on children.”

The increased risk of suicidal ideation and suicide is low, while the risk of stopping antiepileptic drugs is substantial, including death, she explained. The death rate among patients who take their medications is 4.35 per 1,000, compared with 20 per 1,000 among nonadherent patients.

The FDA panel voted against a black-box warning (Psychiatric News, August 15, 2008), and the American Epileptic Society is working on a statement for monitoring recommendations that were due out last month.

Besides any contribution to the risk of suicide, different antiepileptics have a number of other side effects of which psychiatrists should be aware.

“Most antiepileptic drugs have side effects that could be psychiatric, but terms like ‘cognitive effects’ are too broad,” said Plioplys.

For instance, long-term use of phenobarbitol can lead to a 10-point decline in IQ scores. Topiramate can induce psychomotor slowing, memory decline, and decreased verbal learning. Levetiracetan can cause acute psychosis. The combination of topiramate and valproic acid can produce hyperammonemia.

Valproic acid can also cause hemorrhagic pancreatitis, which can progress rapidly from mild nausea to death. The hepatic fatality rate is higher in children than in adults and can occur at the start of treatment or several years later. With valproic acid, clinicians should monitor platelet count and coagulation parameters.

There have been reports of metabolic side effects for topiramate, including cognitive slowing, oligohidrosis, kidney stones in adults (hydrate to reduce stone formation), and increased body temperature.

Risk factors for lamotrigine are young age, co-administration with valproic acid, and high doses. A life-threatening rash may appear between two and eight weeks, and discontinuation may not halt it.

“This is a hypersensitivity reaction that can lead to multiorgan failure, not just a rash.”

Carbamazepine may also cause a hypersensitivity reaction, manifesting early as fever, pruritis, or arthralgia—with or without rash—and should be treated as a medical emergency.

Drugs for ADHD

The most common side effect of stimulant drugs is anorexia, occurring in 13 percent to 22 percent of cases, although the effect decreases over time, said Josephine Elia, M.D., of Children's Hospital of Philadelphia.

“The effect is statistically significant but is not likely to be clinically significant,” said Elia. Methodological problems complicate study of stimulant side effects.

“Most subjects are not drug naive at the start of trials, there is no differentiation between obese and normal subjects, and patients spend varying lengths of time on drugs,” she said.

Some drugs require specific care in monitoring. Atomoxetine calls for a liver enzyme test at baseline and one or two more after treatment begins. No catastrophic hepatotoxic events have been reported, but elevated enzyme levels call for stopping the drug.

An FDA warning in 2005 pointed out risk of cardiac abnormalities in patients taking ADHD drugs.

One retrospective study found that 1.8 percent of young victims of sudden unexplained death were taking stimulants, versus 0.4 percent of matched controls who died in car crashes.

“This supports an increased risk of cardiac effects with stimulants, or possibly for ADHD itself,” said Elia. Stimulants do increase the heart rate 3 BPM to 10 BPM and raise blood pressure 3 mmHg to 4 mmHg, statistically, but not necessarily clinically, significant elevations. Outliers whose heart rate increases by at least 25 BPM (or reaches 110 BPM) or whose blood pressure rises above 130 mmHg should be monitored, she advised.

Atypical Antipsychotics

“Side effects of atypical antipsychotics are due to a complex interaction of pharmacogenetics, drug type, dosage, polypharmacy, and socioeconomic issues,” said Chadi Calarge, M.D., an assistant professor of psychiatry at the University of Iowa.

The two primary concerns with atypical antipsychotics are weight gain and hyperprolactinemia.

“Don't look at children's absolute weight,” he said. “Children are supposed to grow. Consider weight gain in relation to expected changes in height or weight.”

Higher baseline weight seems to lead to more weight gain on the drugs, and adult studies suggest that initial weight gain predicts later weight gain.

Calarge suggests collecting comprehensive personal and family histories and baseline data on weight, BMI, lipids, fasting blood sugar, and monitoring them quarterly, to start.

Hyperprolactinemia risk appears to be greatest with risperidone and with olanzapine over time. Aripiprazole may decrease prolactin over time.

Gynecomastia or even galactorrhea are occasional side effects, but can occur in boys going through puberty and so are hard to detect, especially in obese patients.

“Always ask and don't depend on self-reporting,” he said.

Clinical Trials

Deciding how to record side effects of medications in clinical trials is more complicated than it looks at first glance, said Graham Emslie, M.D., a professor of psychiatry and the Charles E. and Sarah M. Seay Chair in Child Psychiatry at the University of Texas Southwestern Medical Center in Dallas.

Just asking patients is not enough. How investigators ask affects how many events get chalked up, said Emslie, speaking on the same program.

Trial protocols can be neutral and just ask, in effect, “What's up?,” hoping the patient will offer useful data. Alternatively, researchers can offer a checklist of symptoms to see what resonates or use system reviews, which elicit more events than spontaneous reports do.

Definitions count, too, said Emslie. “How are criteria set? Did patients simply have symptoms at some point or have they been serious enough to treat?”

Suicidality issues in antidepressant trials raised additional questions for researchers, he said. “What do you measure? Credible intent to kill self? Or self-injury not related to suicide attempts? Or not any actual injury, but potential injury?”

The FDA's call for closer monitoring of patients starting out on SSRIs also can alter the rate of adverse events, he said.

“If you monitor more closely, you will get more adverse events,” he said, although there is no evidence of more frequent visits following prescription of antidepressants. “Seeing and talking with a patient would be better than measuring.”