January 14 (HealthDay News) -- A new study warns that the second generation of antipsychotic drugs, used to treat conditions ranging from schizophrenia to anxiety, put patients at higher risk of sudden death due to cardiac arrest.

The odds of a heart problem are low, and specialists said that the drugs are appropriate for certain patients. Still, doctors, families and patients should be cautious, said study lead author Wayne Ray, director of the Vanderbilt University School of Medicine's Division of Pharmacoepidemiology.

"If they're being used for schizophrenia, consider a cardiology evaluation. If you're considering using them for bipolar disorder, think about using another alternative drug first," Ray said.

And patients should rarely, if ever, take the drugs to treat other conditions, he said.

At issue are newer antipsychotic drugs -- clozapine (Clozaril), quetiapine (Seroquel), olanzapine (Zyprexa) and risperidone (Risperdal).

Federal health officials have approved the use of the drugs to treat schizophrenia and bipolar disorder, Ray said.

Doctors also prescribe them for so-called "off-label" uses to treat conditions such as anxiety, attention deficit disorder in children and dementia in the elderly.

"For schizophrenics, they work pretty well. They're pretty much the only alternative," Ray said. But other drugs offer alternatives for bipolar patients, he said.

For the new study, published in the Jan. 15 issue of the New England Journal of Medicine, Ray and his colleagues expanded on earlier research that suggested the newer drugs disrupt the heart's rhythm. The researchers examined the medical records of 44,218 patients who used the older antipsychotic drugs and 46,089 patients who used the newer ones. All the patients lived in Tennessee and were recipients of Medicaid, the government-sponsored insurance program that serves low-income people.

The researchers also looked at the records of 186,600 people who didn't use antipsychotic drugs.

They found that users of the newer drugs were 2.26 times more likely to suffer from sudden cardiac death than those not on the medications. Those who used the older drugs were 1.99 times more likely to die versus those not taking the medications.

Patients who took the highest doses were at the highest risk. Overall, the patients had a three-in-1,000 risk of sudden cardiac death a year, Ray said.

The numbers may seem low, but they're significant, Ray said. "If I were talking to a friend or family member, I'd advise them to avoid [the drugs] if possible."

The drugs appear to cause problems by disrupting potassium in the heart, causing its electrical rhythm to fail, Ray said.

Dr. Sebastian Schneeweiss, an associate professor of medicine and epidemiology at the Harvard School of Public Health, co-wrote an accompanying commentary in the journal. He said that, considering the risk and lack of evidence that the drugs are useful beyond limited cases, doctors should "sharply" reduce their use to treat conditions other than acute psychosis and schizophrenia.

SOURCES: Wayne Ray, Ph.D., director, Division of Pharmacoepidemiology, Vanderbilt University School of Medicine, Nashville; Sebastian Schneeweiss, M.D., associate professor of medicine and epidemiology, Harvard School of Public Health, Boston; Jan. 15, 2009, New England Journal of Medicine