SNRIs and Benzodiazepines Have Greatest Effect
By Annette M. Boyle
IOWA CITY, IA—Post-traumatic stress disorder (PTSD) appears to increase the risk of dementia in veterans but exactly why remains unclear: Is the increased risk attributable to frequent comorbidities such as diabetes and traumatic brain injury or to medications used to treat PTSD or to other factors?
Research by investigators at the Iowa City VA Health System confirmed some expected associations—and turned up some surprising connections between common medications and elevated dementia risk, even in patients without PTSD.
The study, published in the Journal of the American Geriatrics Society, “went a step beyond prior research that has found an association between PTSD and the onset of dementia,” explained co-author Thad Abrams, MD, MS, of the Iowa City VA Health System and the University of Iowa. “Our study confirmed that association, with findings that were similar in both direction and magnitude.”1
The study tracked 417,172 veterans over age 55 who received care through the VA for nine years and had at least one clinical encounter every two years from 2004-2012. At the start, 5.4% of the veterans had a diagnosis of PTSD and none had a diagnosis of dementia or cognitive impairment.
Previous studies found that PTSD increased the risk of dementia by 52%-77%. Depending on the adjustments made, the Iowa City study showed a comparable correlation, with PTSD doubling the risk of dementia diagnosis over nine years in an unadjusted analysis and increasing it by 36% when adjusted for medical and psychiatric comorbidities, demographics and clinic utilization.
Examining the impact of various medications provided insights not detected in previous studies. The team “noted several interactions between PTSD and various medication on the association with dementia,” Abrams told U.S. Medicine. Overall, veterans with PTSD who took none of the antidepressants or antipsychotics commonly prescribed for the condition had a 55% higher risk of dementia than those who had neither PTSD nor took those any of those medications.
The study found that patients with PTSD who have received selective serotonin reuptake inhibitors (SSRIs), newer antidepressants and atypical antipsychotics had a substantially increased risk of dementia compared to PTSD patients who were not prescribed those three classes of medications and to patients without PTSD who do not use those classes of medications. The results “suggest that the associations between PTSD diagnosis and the risk of dementia diagnosis varies significantly according to receipt of these medication classes,” the authors said.
Some of the drugs had a stronger association with dementia. “The associations between dementia and SSRIs were effect-modified by presence of PTSD,” Abrams said. Veterans with PTSD who had not been prescribed SSRIs had a 55% increased risk of dementia compared to veterans without PTSD who did not take SSRIs, but veterans with PTSD who had been prescribed SSRIs had twice the risk of dementia compared to veterans without PTSD who had not been prescribed SSRIs. A similar association was found with novel antidepressants.
Taking atypical antipsychotics, however, more than doubled the risk of dementia in patients with PTSD, while the combination of PTSD and atypical antipsychotics increased a veteran’s risk fourfold compared to veterans with no PTSD and no antipsychotics.
Of the psychotropic medications analyzed, “only selective norepinephrine reuptake inhibitors (SSRIs) and benzodiazepines (BZAs) appeared to have independent association with dementia,” Abrams said. “This is not unexpected for BZAs but somewhat surprising for the SNRIs.” Both classes of medication were associated with an approximately 35% increase in dementia risk regardless of PTSD diagnosis.
Based on previous research concerning dementia and benzodiazepines, “I feel there is likely to be more of a contribution from BZAs to dementia pathology,” Abrams said. “However, the SNRI finding was somewhat unexpected and I believe it to be more likely related to treatment of other mental or medical health comorbidities that are on the pathway of dementia.”
As dementia often first presents as depression, Abrams urged providers to consider this diagnosis, particularly when treating older veterans with PTSD. In addition, he encouraged clinicians to regularly conduct memory screens in patients with PTSD to pick up signs or symptoms of dementia early in its development.
He cautioned that the “study is a very initial examination of the complex relationship between PTSD and dementia and utilizes observational data to examine for associations across large populations.” While he did not recommend that other physicians rely solely on this observational data when making prescribing decisions for individual patients, he did advocate considering other options for treatment.
“We are learning that evidenced-based psychotherapy practices are stronger and more reliable treatments for PTSD relative to medication-based therapies,” he said. In addition, providers should “follow the PTSD guidelines that are being released right now which support the use of psychotherapy first over medications,” Abrams added.
Still, the authors of the study noted that an interaction between the psychotropic medications studied and PTSD could increase the risk of dementia. In addition, “our analyses showed the risk for dementia diagnosis significantly increased with increasing number of psychotropic medication classes perhaps reflective of PTSD severity.”
“Performing observational data based research is technically challenging and one tends to observe findings that were not initially anticipated,” said Abrams. The results may imply a causal relationship or they may not. Either way, he said, “the inclusion of psychotropic medications in research examining these relationships is important and further efforts should pursue this line of inquiry to either confirm or deny these initial findings.”
- Mawanda F, Wallace RB, McCoy K, Abrams TE. PTSD, Psychotropic Medication Use, and the Risk of Dementia Among US Veterans: A Retrospective Cohort Study. J Am Geriatr Soc. 2017 May;65(5):1043-1050. doi: 10.1111/jgs.14756. Epub 2017 Feb 8.