Questions Remain on Cause of the Mortality Rise
By Brenda L. Mooney
PHILADELPHIA — Parkinson’s disease patients prescribed antipsychotics are significantly more likely to die in the short-term, according to a new veterans study.
As a result, the therapy should only be used as a last resort in those patients with dementia, argued study authors from the Philadelphia and Ann Arbor VAMCs, the Perelman School of Medicine at the University of Pennsylvania and the University of Michigan Medical School.
“I think that antipsychotic drugs should not be prescribed to Parkinson’s patients without careful consideration,” senior author Daniel Weintraub, MD, of the Philadelphia VAMC, emphasized.
This is not the first study suggesting that antipsychotics might do more harm than good. Since the early 2000s, researchers have linked increased mortality with antipsychotic use among patients who have dementia in the general population, and, since 2005, Food and Drug Administration “black box” warnings have been required on antipsychotic drug packaging, warning of the increased risk of death when these drugs are used in dementia patients.
The study published online recently by JAMA Neurology took the research a step further, seeking to determine whether antipsychotic drug use is associated with higher mortality not only in Parkinson’s dementia patients, but in all Parkinson’s disease patients.1
Although usually associated with dementia and later-stage disease, psychosis can occur even in the early stages of illness and in the absence of dementia, said Weintraub. “It happens not uncommonly earlier in the course of the illness.”
Parkinson’s disease (PD) is commonly accompanied by a range of nonmotor symptoms, including psychosis and dementia, according to the article. In fact, psychosis occurs in as many as 60% and dementia in as many as 80% of patients with long-term PD, with dementia a known correlate of psychosis in the disease.
In light of those statistics, the authors reported, use of antipsychotics (AP) in PD is widespread. They cited a study finding that about one-third of patients with newly-diagnosed PD were prescribed an AP within seven years and another finding that a six-year cumulative probability of initiating AP treatment was 50%. A specific VA study that looked at 2,600 patients with PD and psychosis in 2008 found that 50% were prescribed an AP during a one-year period, with dementia significantly associated with increased antipsychotic use.2
For the study, investigators analyzed about 15,000 patient records in the VA database to determine that Parkinson’s patients who began using antipsychotic drugs were more than twice as likely to die during the following six months, compared with a matched set of Parkinson’s patients who did not use the drugs.
A group of 7,877 veterans with Parkinson’s disease who were prescribed antipsychotic drugs at any time during 1999-2010 were compared to a similar control group who did not use antipsychotic drugs. Results indicated that, within the 180 days after they first took antipsychotic drugs, patients in the intervention group had 2.35 times the mortality of the nonusers.
The risk appeared to vary by drug, according to the report, with an increase of 2.16 times for quetiapine fumarate compared with nontreatment, 2.46 for risperidone, 2.79 for olanzapine and 5.08 for haloperidol. About a 50% greater relative mortality risk was identified with first-generation “typical” antipsychotics, such as haloperidol, compared to newer “atypical” antipsychotics such as risperidone and quetiapine, the study notes.
The study pointed out that antipsychotic drugs have a variety of potential side effects, including reduced alertness, increased risks of diabetes and heart disease, decreased blood pressure, and, with longer-term use, even movement disorders that can resemble those seen in PD.
Mortality Cause Unclear
Initial FDA warnings were based on evidence of increased strokes among antipsychotic users, but Weintraub, who also is on the faculty of the Perelman School of Medicine at the University of Pennsylvania, said it is not clear why the drugs are linked to higher mortality in certain patient groups.
“In this study, we looked at the dataset for clues,” said Weintraub, “but the most common cause of death listed was ‘Parkinson’s disease’ — so there really wasn’t anything that pointed to a specific cause or mechanism.”
A follow-up study will examine the same VA database, looking not at risk of death but at disease diagnoses, injuries and other new episodes of ill-health among PD patients taking antipsychotic drugs, comparing them with the same matched controls, he said.
Meanwhile, Weintraub recommended that neurologists and other physicians should take a hard look at other alternatives before prescribing antipsychotics to PD patients. He suggested they might treat any comorbid medical conditions associated with psychosis, reduce the dosage of dopamine-replacement therapies or attempt to and simply manage the psychosis without antipsychotics.
“Antipsychotics should be used in these patients only when the psychosis is of clinical significance, and patients probably should not be left on these drugs long-term without re-evaluation,” he emphasized.
Weintraub’s level of concern about use of antipsychotics in Parkinson’s patients is not universal, however.
In a commentary accompanying the JAMA Neurology article, Mark S. Baron, MD, of the Hunter Holmes McGuire VAMC and Virginia Commonwealth University Health System, both in Richmond, pointed out the limitations of the study, noting, “as in other related studies, we are uncertain whether the underlying psychosis or the treatment itself poses the risk for mortality.”3
Baron noted that the “studies in nonelderly individuals without dementia do not find that APs pose a universal risk for death. Instead, at least atypical AP use has been associated with reduced mortality in this population,” adding, “Further, in conflict with the FDA’s conclusions, numerous newer studies suggest that AP use in elderly patients with dementia may not in fact pose an increased risk for mortality.”
At least one study, he pointed out, “concluded that increased mortality in these populations correlates with the extent of the underlying dementia and psychiatric symptoms independently of the use of conventional and atypical APs.”
In terms of Weintraub’s study, Baron said the level of risk is in line with that regularly associated with use by elderly patients with dementia but not Parkinson’s disease. On the other hand, he argued that the research is incomplete, leading to the question of “whether the underlying psychosis or the treatment itself poses the risk for mortality.”
“Faced with a patient with PD with uncontrollable hallucinations or, for that matter, an elderly patient with dementia and delirium, healthcare professionals are confronted with a difficult situation, often with no effective and clearly safe approaches,” he wrote. “Certainly, the healthcare professional must first look diligently for possible contributory causes, including removing any potential offending agents, especially any anti-Parkinsonians besides levodopa. If unsuccessful, ideally the dose of levodopa should then be lowered cautiously and a plan discussed to increase the dose quickly, if the desired effect is not achieved and the person is rendered physically worse. Published studies do not adequately control for potential contributions of lowering Parkinsonian dosages to increased mortality. Trials of such agents as cholinesterase inhibitors can also be tried.”
If an antipsychotic is used, Baron advised, “atypical agents with the least potential risk for mortality and the lowest associated risk for worsening Parkinsonism should be prescribed. The therapy should be started at low doses and titrated slowly to the lowest effective dose.”
The boxed warning makes families resistant to considering introducing an antipsychotic, he further pointed out, but suggested that “once newer non-AP agents are available and similarly assessed for the risk for death, we may be able to better infer the true risk of using APs in elderly patients with dementia, with or without PD. Ultimately, the FDA may need to reassess the evidence from published studies and to consider the value of supporting a large well-designed prospective study.”
1 Weintraub D, Chiang C, Kim HM, Wilkinson J, Marras C, Stanislawski B, Mamikonyan E, Kales HC. Association of Antipsychotic Use With Mortality Risk in Patients With Parkinson Disease. JAMA Neurol. 2016 Mar 21. doi: 10.1001/jamaneurol.2016.0031. [Epub ahead of print] PubMed PMID: 26999262.2 Weintraub D, Chen P, Ignacio RV, Mamikonyan E, Kales HC. Patterns and trends in antipsychotic prescribing for Parkinson disease psychosis. Arch Neurol. 2011 Jul;68(7):899-904. doi: 10.1001/archneurol.2011.139. PubMed PMID: 21747029; PubMed Central PMCID: PMC3141727. 3 Baron MS. Antipsychotics and Increased Mortality: Are We Sure? JAMA Neurol. 2016 Mar 21. doi: 10.1001/jamaneurol.2016.0213. [Epub ahead of print] PubMed PMID: 26999765.