To The Editor:
We read with great interest the article "Prevalence of Fat Embolism Following Bilateral Simultaneous and Unilateral Total Hip Arthroplasty Performed with or without Cement. A Prospective, Randomized Clinical Study" (2002;84:1372-79), by Kim et al. We would like to congratulate the authors on a very thorough and well-conducted study.
The authors reported on four patients who had neurological manifestations on the first postoperative day. They suggested that all four of the affected patients had blood samples taken on the first postoperative day that were positive for bone-marrow cells. They hypothesized that bone-marrow-cell embolization caused biochemical and/or biomechanical changes that could explain the encephalopathy displayed by this subgroup. They also reported that all patients with bone-marrow-cell embolization had significantly lower arterial oxygen tensions and oxygen saturations than did patients without bone-marrow-cell emboli. Therefore, hypoxemia appeared to be the most likely cause for the neurological manifestations.
First, did the authors consider screening their study patients for a patent foramen ovale? The prevalence of patent foramen ovale in a normal population has been reported to range from 17% to 32% of healthy individuals
1 . This abnormality might be another potential route for emboli to reach the systemic circulation and end organs. Approximately 50% of their study population had evidence of fat emboli in the right atrium; a substantial proportion of the patients might have had a patent foramen ovale that allowed systemic embolization and subsequent cerebral embolization, resulting in cognitive dysfunction that may not have manifested as floridly as encephalopathy and confusion.
Second, did any of these four patients have evidence of fat emboli as well? A total of twenty-five patients had bone-marrow-cell embolization, but only four displayed neurological symptoms. Is it possible, therefore, that the remaining twenty-one patients had manifestations too subtle to pick up with standard clinical neurological examinations?
Milder forms of cognitive dysfunction might be better explained by the work carried out by Byrick et al.
2 and others
3,4 who have shown repeatedly that intravascular fat has been found in all brain, kidney, and heart specimens obtained after intramedullary procedures in canine models. The route for emboli to the systemic circulation was reported to be transpulmonary, as none of the dogs had a patent foramen ovale.
To our knowledge, a battery of validated neuropsychological tests is the only reliable method of demonstrating postoperative cognitive dysfunction
5 ; thus, do the authors agree that even the patients who exclusively had fat embolization might have suffered mild cognitive dysfunction that may have been too subtle to pick up by standard clinical means? If all of the patients had been tested by formal neuropsychological means, is there a possibility that cognitive dysfunction would have been detected in a subgroup of patients positive for patent foramen ovale and fat emboli, in the presence of normal perioperative arterial oxygen tensions and saturations as were reported for the group with fat emboli only?
Furthermore, do the authors agree that, although the four patients who had florid neurological manifestations also had hypoxemia, their cognitive dysfunction might have been a result of fat or bone-marrow-cell cerebral embolization through a patent foramen ovale or through transpulmonary passage? Would they also agree, therefore, that a larger proportion of their study sample may have suffered milder cognitive dysfunction, attributable to the same embolic routes?
Whether a relationship exists between this subgroup and unilateral or bilateral hip arthroplasty and hip arthroplasty with or without cement remains unclear.
Y.-H. Kim, S.-W. Oh, and J.-S. Kim reply:
We are pleased to respond to Dr. Patel's commentary on our article. First, we did not consider screening our study patients for a patent foramen ovale. We agree that this might be another potential route for emboli to reach the circulation and end organs. We monitored right atrial pressure, and yet we found no abnormally increased right atrial pressure. Therefore, we believe that the possibility of systemic embolization through a patent foramen ovale is slim.
Second, four patients with bone-marrow emboli had no evidence of fat emboli. However, it is possible that the remaining twenty-one patients had manifestations too subtle to pick up with standard clinical neurological examinations. Even if milder forms of cognitive dysfunction were not detected with standard clinical neurological examinations, this transient mild neurological dysfunction would not be a relevant clinical problem.