To the Editor:
Avidan et al. 1have taken on the difficult task of looking through the vast Alzheimer's Disease Research Center database to clarify the effect of noncardiac surgery on the progression of Alzheimer's disease and age-matched controls. Alzheimer's disease is a long-term condition of progressive cognitive deterioration of unknown etiology, with widely accepted pathologic markers.2It is not surprising that single episodes of various medical illnesses, surgery, and anesthetic techniques cannot be shown to alter its course, as the authors rightly conclude. However, we think that the lack of observed effect, reported in the 214 nondemented participants, warrants further scrutiny.
As in previous studies,3,4the authors have based their conclusions on a composite score, derived from a battery of neuropsychological tests. In one previous study to which the authors refer, tests were selected by the investigators, based on their “appropriateness.” The cognitive tests used in the present study are those selected for tracking the functional changes of Alzheimer's disease. In either case, tests to identify or dismiss any cognitive deficit resulting from anesthesia or surgery have not been systematically sought.
Testing in the current study has been performed at different intervals, after a variety of operations or illnesses, and different numbers of tests were performed on the participants, all included in the calculated slopes, which are the basis for their conclusions. The authors also agree that the testing was skewed, in that a relatively small number of participants underwent multiple testing. This pool of heterogeneous data limits the impact of the study. Given the wide range of illnesses, operations, follow-up intervals, numbers of tests, etc. , we wonder why the authors have chosen not to provide the data for individual tests, follow-up intervals, or data plots to support their conclusions.
In addition, the collection of postevent data differs with mean values of 1.2 or 2.1 yr (illness and surgery groups, respectively). Is it acceptable for a patient to have postoperative cognitive dysfunction for a median of 2.1 yr after surgery, provided they (eventually) resume their existing level of decline in cognitive function? It also leads us to question the authors' headline conclusion that “The decision to proceed with surgery in elderly people, including those with early Alzheimer's disease, may be made without factoring in the spectre of persistent cognitive deterioration.” This is not only an unacceptable conclusion but may also hinder future interest and research in an area that the authors themselves accept is of great importance!
This is a retrospective report, and the authors have not designed the study to answer their question directly. Collection of data at intervals to reproduce the transient cognitive defect found in the early postoperative phase (reported by Moller et al. ), followed by the absence of a long-term deficit, would have been compelling support for the authors' hypothesis. We hope that the authors will use the resources of their Alzheimer's Disease Centre to perform a prospective study, designed to delineate the effect of anesthesia and surgery on cognitive function in this age group.
*King's College Hospital, London, United Kingdom. padma.nainar@nhs.net
