Post-Op Delirium Linked to Accelerated Cognitive Decline

Postoperative delirium in older adults was associated with 40% faster cognitive decline, data from a prospective cohort study showed.

Among older adults who developed delirium after major elective surgery, cognitive decline over 72 months occurred at a rate of 0.14 population standard deviation (SD) units per year, reported Zachary Kunicki, PhD, MS, MPH, of Brown University in Providence, Rhode Island, and co-authors.

This was significantly faster than the long-term cognitive decline of 0.10 population SD units per year in people who either didn't develop delirium or didn't undergo surgery, the researchers wrote in JAMA Internal Medicine.

Whether delirium actually caused faster cognitive decline or was simply a marker of people with preclinical brain disease is unknown.

"This study has the longest follow-up period of any study examining persons with delirium following surgery," said co-author Sharon Inouye, MD, MPH, of Harvard Medical School and the Aging Brain Center in Boston, in a statement.

"While future studies are needed, this study raises the possibility that delirium may predispose to permanent cognitive decline and potentially dementia," Inouye noted. "This highlights the importance of delirium prevention to preserve brain health in older adults who undergo surgery."

In both surgical and nonsurgical patients, delirium has been linked with subsequent cognitive decline. A recent meta-analysis showed that people who experienced an episode of delirium were more than twice as likely to show long-term cognitive decline as those without delirium.

Kunicki and colleagues followed 560 community-dwelling older adults in the ongoing SAGES study that began in 2010. Mean age was 76.7, and 58% of the cohort was female; at baseline all participants were dementia-free. All participants had elective surgery -- mostly orthopedic procedures -- with an anticipated hospital stay of 3 days or longer.

During hospitalization, delirium was assessed daily with the Confusion Assessment Method (CAM) and supported by medical record review. Overall, 134 people (24%) developed postoperative delirium.

Cognitive performance was assessed with a battery of 11 tests preoperatively and at multiple timepoints after surgery for up to 6 years. The tests measured attention, memory, language, and executive function. The researchers evaluated an additional 119 participants as a nonsurgical comparison group to quantify learning associations with cognitive retesting.

Longitudinal cognitive change was based on general cognitive performance (GCP), a composite measure scaled so that 10 GCP points were equivalent to 1 population standard deviation. GCP scores were corrected for retest effects.

In all patients, the average cognitive changes seen after surgery included an abrupt decline 1 month after surgery, an increase in cognitive performance 2 months after surgery, a stable period 6 to 30 months after surgery, and a steady decline 3 to 6 years after surgery. Long-term cognitive change occurred at a pace of about -1.0 GCP units (95% CI -1.1 to -0.9) per year.

Patients who experienced delirium after surgery had a sharper drop at 1 month, greater recovery at 2 months, and faster decline in all time periods from 6 months to 6 years. After adjusting for practice and recovery effects along with confounders including age, sex, race, comorbidities, depression, and preoperative decline, long-term cognitive change for patients with delirium was an additional -0.4 GCP units (95% CI -0.1 to -0.7), for a total of -1.4 units per year.

"In overall comparisons, the group who developed delirium (compared with those who did not) showed a significantly steeper slope of cognitive decline that approximated the slope seen among individuals who receive a diagnosis of Alzheimer's disease within 5 years," Kunicki and co-authors observed.

The study had several limitations, the researchers acknowledged. The sample was highly educated and predominantly white, and results may not apply to other groups. The analysis did not control for post-delirium illnesses or medications that may have influenced long-term cognitive decline.

"Importantly, information on amyloid status from cerebrospinal fluid biomarkers or amyloid imaging was not available in this cohort; thus, the associations of these variables with cognitive trajectory could not be evaluated," Kunicki and colleagues added.

Comment