More Evidence Atypical Antipsychotic Use May Boost Kids' Cardiovascular Risk

Caroline Cassels

November 10, 2009

November 13, 2009 [Updated with comment] — More research appears to confirm recent results of a large, prospective cohort study that the use of atypical antipsychotics (AAPs) increases the risk of significant weight gain and varied metabolic changes in children and adolescents with mental illness and behavioral disturbances.

Two new studies presented here for the first time at the American Academy of Child & Adolescent Psychiatry 56th Annual Meeting confirm findings by Christoph U. Correll, MD, and colleagues published in the October 28 issue of the Journal of the American Medical Association that these medications cause weight gain and are associated with varied adverse metabolic changes.

In the first study, investigators from the University of British Columbia, Children's and Women's Health Centre in Vancouver, Canada, found the overall prevalence of metabolic syndrome (MetS) was 27% in AAP-treated patients vs 2.9% in AAP-naive youth. In addition, the researchers found higher rates of all features of MetS in AAP-treated youth. The study also showed significantly increased rates of other metabolic parameters in AAP-treated patients.

Rates of MetS Features in AAP-Treated and AAP-Naive Patients

MetS Features Percentage of AAP-Treated Patients Percentage of AAP-Naive Patients
Obese 32.7 16.3
Overweight 23.6 12
Elevated waist circumference 49.1 17.9
Fasting hypertriglyceridemia 42.6 22.4
Elevated fasting glucose level 16.1 2.6
Elevated blood pressure 53.7 17.6

"One of the most startling findings was the prevalence of metabolic syndrome in kids treated with atypicals vs the non-treated kids. I think this is the first time such a high prevalence has been reported," principal investigator Dina Panagiotopoulos, MD, told Medscape Psychiatry in an interview.

Growing Body of Evidence

According to Dr. Panagiotopoulos, a growing body of evidence in adults demonstrating that AAPs cause significant weight gain, hyperlipidemia, and insulin resistance has raised concerns among the medical community in general, and the psychiatric community in particular, about whether these drugs may increase the risk of premature cardiovascular disease in children and adolescents.

It is estimated that approximately 20% of youth in the United States have a mental illness and many of them will be treated with an AAP, said Dr. Panagiotopoulos.

"We decided to undertake this research based on observations of sudden-onset diabetes in some of our young patients who were taking these medications — particularly quetiapine," Dr. Panagiotopoulos said.

Dr. Panagiotopoulos added that based on guidelines in adults issued by the American Diabetes Association, the American Psychiatric Association, and other groups in 2004 that recommended regular monitoring of metabolic parameters in adults taking AAPs, her research team set out to develop a systematic approach to look at the metabolic side effects in children and adolescents. There are currently no such guidelines for children taking AAPs.

"We knew this had never been done in a consistent or systematic way and so we implemented a metabolic monitoring protocol. The reason we are so concerned is not just because of the diabetes risk, but we know that some of these other risk factors, such as dyslipidemia and elevated waist circumference, can track into adulthood and increase the risk for cardiovascular disease and other complications," said Dr. Panagiotopoulos.

Metabolic Monitoring Protocol

To assess the metabolic effects of AAPs in youth, the investigators compared AAP-treated and AAP-naive youth using the metabolic monitoring protocol they developed. The pediatric definition of MetS consisted of at least 3 of the following parameters:

  • Triglycerides ≥ 1.24 mmol/L (110 mg/dL)

  • HDL ≤ 1.03 mmol/L (40 mg/dL)

  • Waist circumference ≥ 90th percentile for age and sex

  • Systolic or diastolic blood pressure ≥ 90th percentile for age, sex, and height

  • Fasting glucose ≥ 5.6 mmol/L (100 mg/dL)

The study included 176 patients admitted for psychosis or bipolar disorder to a short-stay, inpatient psychiatric unit from January to December 2008.

Among the participants, 66 were taking AAPs (57.5% male; mean ± SD age, 12.8 ± 2.7 years) and 110 (54.5% male; mean ± SD age, 14.3 ± 2.4 years) were AAP naive on admission.

Of the total study group 27 were taking risperidone, 23 were taking quetiapine, 13 were taking olanzapine, and 2 were taking both risperidone and quetiapine.

The study was not powered to detect differences among agents. However, Dr. Panagiotopoulos said the research team will continue to collect data and then conduct further analyses in a larger sample size to answer this and other research questions.

"Our initial analysis suggests that olanzapine may be more likely to confer additional risk. We conducted a regression analysis looking at additional proposed independent risk factors, including mental illness itself, ethnicity, or the addition of other [psychotropic] medications, such as antidepressants or mood stabilizers, but there was no evidence that this was the case."

However, she added, there was a significant trend toward male sex being an independent risk factor. "We're going to have to see whether, with an increased sample size, this finding holds true," said Dr. Panagiotopoulos.

"Overall," she added, "the study findings were pretty consistent with Dr. Correll's recent study in JAMA and so there really weren't any surprises here."

Surprise Finding?

In contrast, a second study led by Margaret Weiss, MD, PhD, who was also a coinvestigator on the first study, produced a surprising finding.

The cross-sectional study examined the metabolic effects of very low-dose AAPs in children with attention-deficit/hyperactivity disorder (ADHD) with comorbid disruptive behavior disorders. They found that despite very low doses and low body mass index children taking AAPs also presented with metabolic disturbances.

"Because these children are always very thin, it was assumed that there would be no metabolic findings, but actually what we found was that more than 16% had full metabolic syndrome and what was most interesting was that even though they were very thin, their waste circumference grew — so it turned out it was waist circumference and not weight that was a predictor of metabolic syndrome," Dr. Weiss told Medscape Psychiatry.

However, Dr. Correll, principal investigator of the JAMA study, said that he does not fully agree with the researchers' interpretation of the results. "The investigators note that the kids had normal to low BMI, but this statement is only true when looking at the regular BMI, which does not take into account the developmentally adjusted BMI norms of the youngsters.

"When you look at the BMI z scores (a measure of body mass adjusted for children's age and gender), these data indicate that the sample was almost 1 full standard deviation above the age- and sex-matched norm," Dr. Correll told Medscape Psychiatry.

In addition, he noted that the study shows that 42% of the patients were obese or overweight, and therefore the increased prevalence of metabolic abnormalities "is not that surprising."

He also pointed out that the cross-sectional design of the study does not provide information about weight and metabolic status before treatment with AAPs and noted that there was no control group for any of the cardiometabolic findings in this study, making it difficult to fully interpret the observed prevalence rates of the "admittedly concerning" metabolic abnormalities.

The study included 194 children being treated for ADHD in a single center. Of these, 23 were taking no medication, 134 were taking ADHD medications alone, and 37 were taking AAPs to treat behavioral problems, including reactivity, rages, and mood dysregulation. Metabolic data were available for 27 patients, all of whom were taking AAPs at the time of the assessment.

The mean age of children in the AAP group was 12.3 years, and most (86.5%) were male. Most patients in the AAP group were taking risperidone, with a mean dose of 0.62 mg. The mean duration of treatment was 313 days.

More than two-thirds of the 27 patients taking AAPs (68.0%) had a waist circumference in the 90th percentile or higher independent of weight gain. In addition, 18.5% had an impaired fasting glucose level, 12.5% had elevated blood pressure, 11.1% had elevated levels of triglycerides, and 16.7% met full criteria for MetS.

Children Taking Low-Dose AAPs Should Be Monitored

In response to Dr. Correll's concerns, Dr. Weiss countered that although Dr. Correll is correct in his observation that the BMI z score of children with ADHD was 0.88, in her opinion this only reinforces the point of this small pilot study, which was to demonstrate that the use of low-dose risperidone in children who were initially thin and who had decreased appetite because of stimulant use raises the same issue of metabolic risk and weight gain as it does in other populations.

"Dr. Correll may want to consider doing a stratified analysis of this population in his study comparing children on stimulants and low-dose risperidone for irritability to children on higher doses of atypicals for treatment of psychosis to determine if there's a difference in the risk profile," said Dr. Weiss.

"In the meantime," she added, "the data are sufficient to indicate that it would be prudent to include laboratory monitoring when using atypicals, even in children on low doses of these agents with no evident overweight or obesity."

Urgent Need for Guidelines

Of concern, said Dr. Weiss, is that off-label use of AAPs to treat disruptive disorders in children with ADHD is common, with 1 in 5 patients in this study receiving these medications.

Further studies are needed to determine whether the widespread use of these drugs in this patient population has long-term negative health consequences and what the therapeutic gains are relative to other, lower-risk interventions, both pharmacologic and nonpharmacologic.

In the meantime, both Dr. Weiss and Dr. Panagiotopoulos believe there is an urgent need for increased metabolic monitoring in young patients taking these medications.

"These are laboratory findings and whether they do in fact represent a risk for long-term cardiovascular risk remains to be seen. The most reasonable response right now is to develop well-thought-out practice guidelines.

"We put out practice guidelines for cardiovascular monitoring for the use of stimulants without any evidence whatsoever that there is a cardiovascular risk. Here, we have unequivocal evidence but no practice guidelines have been adopted, so the vast majority of children taking these medications receive no laboratory monitoring whatsoever," said Dr. Weiss.

Dr. Panagiotopoulos added that there is also a need to increase awareness among psychiatrists of the potential metabolic effects in children taking AAPs and a need to institute standardized metabolic monitoring.

"One of the big challenges here is that mental health care is not 'medicalized.' Most kids just go in for counseling and that's it. There is no infrastructure to support metabolic monitoring, and so I think in light of this growing body of evidence we may have to change the model, so kids can be appropriately monitored over the long-term," she said.

Dr. Panagiotopoulos added that her team is currently developing educational resources about the metabolic effects of AAPs in children and adolescents, including a handbook for pediatric residents, which is currently being tested in a pilot study.

"We want to turn that knowledge into action and change practice, and so we figure by starting with pediatric residents we may be able to help the next generation of children and physicians," she said.

Preventive Strategies "Sorely Needed"

Dr. Correll said the investigators should be complimented for establishing regular monitoring schedules for cardiometabolic risk factors in AAP-treated youth attending their centers.

Despite some of his questions regarding the interpretation of the data in the second study, he noted that these 2 studies provide important evidence that treatment with AAPs is associated with problematic rates of overweight, obesity, and metabolic abnormalities.

In particular, he said, the high and elevated rate of full MetS in the study by Dr. Panagiotopoulos and colleagues raises concerns about the risk for future cardiovascular morbidity and mortality in this patient population.

"Consistent with recently published study results from our group, patients receiving stimulants and atypical antipsychotics together do not seem to be protected from having elevated body weight for their sex and age," Dr. Correll told Medscape Psychiatry.

"The results from both studies emphasize that the selection of antipsychotics needs to be weighed carefully and that the risks of an untreated or suboptimally managed psychiatric disorder need to be balanced against the cardiometabolic risk of antipsychotics that differs across the available agents.

Dr. Correll and Dr. Panagiotopoulos have disclosed no relevant financial relationships.

American Academy of Child & Adolescent Psychiatry (AACAP): Abstracts 3.40 and 4.25. Presented October 29 and October 30, 2009.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.