The Impact of Cerebral Fat Microemboli on Cognitive Function and Magnetic Resonance Diffusion-Weighted Brain Imaging Following Elective Joint Replacement Surgery in the Elderly

Sebastian Koch, M.D.

University of Miami, Miami, FL

Grant Program:

David Mahoney Neuroimaging Program

Funded in:

February 2002, for 3 years

Funding Amount:


Lay Summary

The Impact of Cerebral Fat Microemboli on Cognitive Function

Cognitive decline, such as memory difficulties, may occur in 5-29% of patients after orthopedic surgery.   The reasons for this decline are unknown.  Recently small fat particles have been detected with ultrasound in the brain circulation of patients undergoing skeletal surgery.  It is believed that these fat particles get released from the marrow as the bone is surgically manipulated and subsequently reach the brain.  It is presently unknown if these microemboli cause brain injury.  We here examine the clinical significance of intra-operative cerebral microembolism by assessing cognitive function and brain MR imaging in patients undergoing hip and knee replacement surgery.  We also speculate that a small communication in the heart between the venous and arterial side (venous-arterial shunt), increases cerebral microembolism and adversely affects cognitive function.


The Impact of Cerebral Fat Microemboli on Cognitive Function

Post-operative cognitive dysfunction frequently complicates non-cardiac surgery. In approximately 10-30% of elderly patients, persistent cognitive decline occurs after major general surgery, including orthopedic procedures. Immediately following orthopedic surgery transient confusional states are often noted, and long term neuropsychological dysfunction is found in 5-29% of patients.

The etiology of this neuropsychiatric injury is unknown and appears unrelated to the effects of general anesthesia. Recently transcranial Doppler has shown that cerebral microembolism occurs during skeletal surgery. These microemboli are presumed to be fat emboli. Lipid microembolism contributes to the well known cognitive decline seen after cardiac surgery. It is presently unknown if a similar detrimental effect of lipid microembolism on cognition occurs during orthopedic surgery.

We will examine the clinical importance of intra-operative cerebral microembolism by assessing cognitive function and brain MR imaging in patients undergoing primary joint arthroplasty. We further postulate that in setting a venous-arterial shunt, cerebral microembolism is increased, thereby adversely affecting cognitive function.

Investigator Biographies

Sebastian Koch, M.D.

Assistant Professor of Clinical Neurology, University of Miami



Elective joint replacement surgery in the elderly may lead to postoperative delirium and long-term cognitive decline. The etiology of this neurological injury has remained uncertain. We propose that cerebral fat microembolism is the mechanism responsible for the cogntive impairment seen postoperatively.

We hope to establish a direct correlation between the number of cerebral fat emboli detected intra-operatively with transcranial Doppler ultrasound and the emergence of cognitive dysfunction and diffusion-weighted MR imaging ischemic changes in patients undergoing joint replacement surgery. This will define novel pathophysiological mechanisms in a clinical disorder, where the role of microembolism has received little or no attention. Additionally, we anticipate that a patent foramen ovale will be a risk factor for adverse neurological outcome. This would allow the identification of high-risk patients in a timely manner, preoperatively.

Patients at least 65 years old, requiring knee or hip replacement surgery, will be studied preoperatively with neuropsychological testing and MR diffusion-weighted imaging. All patients will be screened for a patent foramen ovale. Intra-operatively patients will be monitored with transcranial ultrasound for thepresence of fat microemboli. Postoperatively neuropsychological testing and MR diffusion weighted imaging will be repeated to assess the impact of cerebral fat embolism.

We enrolled 24 patients who required elective hip and knee replacement surgery. The mean age of our study population was 74 years, and 14 patients were men. Six patients had hip replacement and 18 had knee replacement. During surgery, all patients had intra-operative microemboli. The mean number and size of emboli was 9.9 ±3.6. We found a large variation in the number of emboli during surgery, which ranged from 1-84.

Cognitive decline was present in 18/24 (75%) at discharge and in 10/22 (45%) at 3 month. Patients with neuropsychiatric dysfunction at discharge had more intra-operative microemboli than patients without dysfunction (11.7± 4.7 vs. 4.7± 1.1) but this difference was not statistically significant. There was no correlation between 3 month cognitive decline and intra-operative microembolism.

We also examined the relationship between emboli and the presence of a venous-arterial shunt. In the 12 patients with a shunt, we found a non-significant trend towards more microemboli (15.4± 6.8 with shunt vs. 4.4±1.0). There were no significant differences in the size of the emboli between the two groups, even though a trend toward larger emboli was noted in patients with a shunt.

All patients completed the MR imaging part of the study. We did not find any new diffusion-weighted imaging lesions post-operatively. Only one patient had an increase in the white matter lesions seen on FLAIR imaging post-operatively as determined by a neuroradiologist who examined the images blinded to study protocol.

We therefore conclude that short and long term cognitive decline is frequently seen after hip and knee replacement surgery. Intra-operative microembolism occurred universally; however, it did not significantly influence post-operative cognitive dysfunction. We also conclude that, despite intra-operative microembolism in all cases, diffusion and FLAIR imaging abnormalities occur infrequently after surgery.

Selected Publications

Koch S., Forteza A., Lavernia C., Romano J.G., Campo N., Bhatia R., and Campo-Bustillo I.  MR imaging after hip and knee replacement. J Neuroimaging. 2007 Oct;17(4):332-5 .

Koch S., Forteza A., Lavernia C., Romano J.G., Campo-Bustillo I., Campo N., and Gold S.  Cerebral fat microembolism and cognitive decline following hip and knee replacement. Stroke. 2007 Mar;38(3):1079-81.

Forteza A.M., Rabinstein A., Koch S., Zych G., Chandar J., Romano J.G., and Campo-Bustillo I.  Endovascular closure of a patent foramen ovale in the fat embolism syndrome: changes in the embolic patterns as detected by transcranial Doppler.  Arch Neurol. 2002 Mar;59(3):455-9.

Koch S., Rabinstein A., Falcone S., and Forteza A.  Diffusion-weighted imaging detects cytotoxic edema in eclampsia.  AJNR Am J Neuroradiol. 2001 Jun-Jul;22(6):1068-70.

Forteza A.M., Koch S., Romano J.G., Zych G., Campo-Bustillo I., Duncan R.C., and Babikian V.L.  Transcranial doppler detection of fat emboli.  Stroke. 1999 Dec;30(12):2687-91 .