Cerebral Amyloid and Cognition after Surgery: Reply

We thank Meng et al. for their letter to the editor regarding our study exploring cerebral β-amyloid deposition and cognition after cardiac surgery.¹  The authors have pointed out several important limitations of our study, which we have already acknowledged in the original publication.

Certainly, postoperative cognitive dysfunction has a complex pathophysiologic basis without a clear etiology in current understanding. Our study was designed to explore one potential mechanism contributing to cognitive decline after cardiac surgery, while adjusting for other well-established predictors. However, as we acknowledged, our sample size was limited and only designed to explore the relationship between global cortical amyloid deposition, using the novel positron emission tomography β-amyloid tracer ¹⁸F-florbetapir, and cognitive change after cardiac surgery. We do not draw any definitive conclusions from our results, but rather present them to add to the discussion in the field of postoperative cognitive decline research and to potentially generate hypotheses and offer methodologies for future investigations. Limiting analyses to subjects with normal amyloid burden, as suggested by Meng et al., also seems pointless, as the effect of abnormal amyloid deposition could not then be assessed.

With regard to the timepoints selected for imaging, we agree with Meng et al. that baseline imaging would have been optimal, but this was unfeasible given funding constraints for this explorative study. As clearly acknowledged in our publication, we were not assessing change in amyloid deposition from presurgery to postsurgery, given that amyloid deposition takes place over a significantly longer time period (see our reference to the article by Landau et al.² ). Again, as we already noted in our discussion, despite the evidence in the literature for a longer time course for amyloid deposition, we cannot exclude the possibility of a change in amyloid deposition from preoperatively to 6 weeks postoperatively.

Finally, Meng et al. note additional limitations of our study, including the failure to address the association of preoperative amyloid status in patients with persistent (long-term postoperative cognitive decline), the impact of major surgery and anesthesia on amyloid burden, and the association of amyloid deposition in cerebral nuclei with cognitive impairment. To address the first two points: again, our study was explorative in nature and not powered to answer questions related to the influence of baseline amyloid burden on long-term postoperative cognitive decline or how surgery and anesthesia may alter cerebral amyloid burden. Comparing our cohort’s 6-week and 1-yr postsurgical amyloid data did reveal an interesting suggestion of accelerated amyloid deposition compared to known longitudinal amyloid deposition rates in nonsurgical patients (from the Alzheimer Disease Neuroimaging Initiative cohort data³ ); however, once again, our sample size is underpowered to definitively answer these and similar questions, as we have already acknowledged. Finally, with respect to the histologic evaluation of amyloid deposition and its correlation with cognitive impairment, this is difficult to perform in living postsurgical patients, which is why we elected to use noninvasive positron emission tomography imaging with a well-studied amyloid tracer.

We look forward to future investigations to help answer these many important questions related to the potential role of cerebral amyloid deposition in cognitive impairment and postoperative cognitive decline.

The authors declare no competing interests.