No link between patent foramen ovale and cognitive dysfunction

Patients with a PFO had a tendency for larger fat emboli, but size and prevalence were not significant.

CHICAGO — In a study of a small number of patients undergoing total joint arthroplasty, about 75% demonstrated signs of cognitive impairment at discharge, and signs of the impairment lasted as long as three months.

The researchers questioned the implications of the findings, however, because they evaluated patients’ cognitive function according to a strict definition of impairment and performed tests while the patients were on medication.

Investigators theorized the increased impairment probably stems from a surgically related cerebral fat embolism. Their results suggest that the presence of a patent foramen ovale (PFO), a congenital anatomical interatrial communication with potential for right-to-left shunt, has some effect on the quantity and size of these emboli.

But the most disconcerting result, according to researchers, was the severity of the impairment at discharge on postoperative day three, which could have implications for performing total joint replacement (TJR) on an outpatient basis, Carlos J. Lavernia, MD, told Orthopedics Today.

At three months postop, 50% of patients studied continued to have cognitive impairment, said Lavernia, who presented results at the Orthopaedic Research Society’s Annual Meeting, here.

Further analysis

The literature reports the postoperative confusion rate in these cases is as high as 20%, and it is typically related to a cerebral fat embolism during TJR. With an expected increase in the TJR procedures done each year in the United States (there are nearly 700,000 currently performed each year), the cost of evaluating the postop confusion can only increase, according to Lavernia.

Still, the clinical significance of the postoperative cognitive decline the researchers measured remains unclear, he told Orthopedics Today.

“This needs further investigation. The cognitive measures we use do not always correlate with actual functional deficits. Also, the published definition of neuropsychological decline fails to establish a connection between measured cognitive decline and decline in actual activities of daily living,” he said.

“Our group feels that a study should be performed to assess the clinical implications of the published threshold for neurocognitive decline.”

PFO of interest

In their study, Lavernia and colleagues at the Orthopaedic Institute at Mercy, Miami, and the University of Miami, analyzed the type of intraoperative emboli that occurred in 23 patients undergoing TJR (57% women; mean age 74.3 ± 1.5 years). They also assessed other complications.

Researchers identified a microembolism in all 23 patients.

The researchers were particularly interested in whether the presence of a PFO influenced who became cognitively impaired. They suggested that this shunt might allow more fat embolism particles to circulate and ultimately reach the brain.

To look for a PFO, they excluded patients with prior histories of stroke, dementia, atrial fibrillation and carotid stenosis. Preoperatively, patients underwent specialized MRI scans to determine who had a suspected venous-to-arterial (V-A) shunt. Intraoperatively, researchers also monitored patients using noninvasive transcranial Doppler (TCD) ultrasound. Lastly, at 72 hours postop, each patient underwent a repeat MRI.

A neuroradiologist reviewed the images and an investigator blinded to the study protocol interpreted the TCD results.

Emboli size, number studied

“Our cohort of patients is clinically doing extremely well. It is well known that quality of life and functional status improve after joint replacement, reflecting improved activities of daily living and pain scores. The quality of life, function, motion and pain level were improved on all of our patients,” Lavernia noted.

However, the researchers identified a V-A shunt in 11 patients; three undergoing hip replacement and eight undergoing knee replacement.

“The mean size of the emboli in the patients that had the shunt was higher, but it wasn’t statistically significant,” Lavernia said. On TCD, the emboli measured about 13 decibels in the V-A group compared to 8.5 in the non-V-A group.

Other complications included one case each of deep venous thrombosis and transient vision, but no cases of acute confusion.

“We did subgroup analyses and found no statistical differences in size or amount of emboli. We did however observe a trend that perhaps with a larger cohort may become significant. At some point, we may be able to close some of these PFOs,” Lavernia said.

For more information:
Lavernia CJ, Koch S, Forteza A, et al. Cognitive dysfunction in arthroplasty surgery: Cerebral fat embolism? #181. Presented at the Orthopaedic Research Society’s Annual Meeting. March 19-22, 2006. Chicago.