No Link Between Anesthesia and Dementia

Pauline Anderson

May 20, 2013

Exposure to general anesthesia during medical procedures after age 45 years is not a risk factor for dementia, results of a new study suggest.

The lack of association between anesthesia and dementia was also quite robust in several of the study's sensitivity analyses exploring factors such as the timing of anesthetic exposure before the dementia diagnosis and the age at the time of anesthetic exposure, researchers report.

"The message from this paper is very simple: Anesthesia is not a risk for dementia," said one of the study authors David S. Knopman, MD, Department of Neurology, Mayo Clinic, Rochester, Minnesota.

However, the study does not settle the controversy surrounding the concept of postoperative cognitive decline (POCD), the researchers caution.

The retrospective, population-based, nested case-control study was published online May 3 in the Mayo Clinic Proceedings.

Anesthesia Exposure

Dr. David S. Knopman

Through the Rochester Epidemiology Project and the Mayo Clinic Alzheimer's Disease Patient Registry, researchers identified residents of Olmsted County Minnesota who were diagnosed with dementia or Alzheimer's disease (AD) between January 1, 1985, and December 31, 1994.

Experienced behavioral neurologists confirmed the diagnosis of dementia, classified the type of dementia, and determined the year of onset. Patients with postoperative confusion who later returned to their normal mental state were not considered demented.

Using the Diagnostic and Statistical Manual, Fourth Edition (DSM-IV), the investigators distinguished AD from other dementias on the basis of available clinical and laboratory data. Although the criteria for AD dementia have recently been revised, the fundamental features supporting the clinical diagnosis of AD are still reflected in DSM-IV, said the authors.

From medical records, researchers gathered information on exposure to general anesthesia as well as the anesthesia agents used, the type of procedure, and the duration of anesthesia.

For each case, the researchers randomly selected a dementia-free sex- and age-matched control from the general pool of Olmsted County residents.

The study included 877 matched case-control sets. Each control was followed up for at least 5 years, during which time 72 developed dementia and were also included as cases. Most cases (83.5%) met criteria for AD, with the remainder meeting criteria for other dementias.

The median age at the time of the first medical record entry was 31 years for cases and 32 years for controls. The median length of the medical record after age 45 years and before the index date was similar for cases (34.3 years) and controls (34.4 years).

Of the 877 cases, 70.1% underwent 1681 operations or procedures performed under general anesthesia after age 45 years and before a formal diagnosis of dementia, with a median duration of anesthesia of 90 minutes. In the control group, 72.5% underwent 1638 operations, with a median duration of anesthesia of 90 minutes.

No Associations

When exposure to anesthesia was assessed as a dichotomous variable (any vs none), anesthesia was not associated with dementia (odds ratio [OR], 0.89; 95% confidence interval [CI], 0.73 - 1.10; P = .27).

There was also no association when anesthesia was quantified as the number of procedures (ORs, 0.87, 0.86, and 1.0 for 1, 2 to 3, and 4 or more exposures, respectively, compared with no exposures; P = .51) or when total cumulative duration of exposure was assessed as a continuous variable (OR, 1.00; 95% CI, 0.99 - 1.01 per 30-minute increase; P = .86).

When the analysis was limited to only cases of AD, the results were similar.

An analysis of the risk according to individual anesthetic agent also did not reveal any association with incident dementia. The findings were about the same in analyses performed by using a model that included the highest level of formal education as a covariate.

A similar lack of association was found in several sensitivity analyses, including assessments of only participants with complete medical records available after age 45 years; of exposures occurring within 5, 10, or 20 years of the index date; of the case-control sets excluding the 72 in which dementia had developed in controls before 1994; and of only exposures to procedures and anesthesia after age 60 years.

On the topic of POCD, the authors noted a previous review concluding that the current evidence on an association between anesthesia/surgery and POCD is not persuasive. That review highlighted several challenges in studying this area, including the lack of diagnostic criteria for POCD and the potential role of preexisting cognitive impairment.

"The entity of postoperative cognitive decline may be real, not because of anesthetic complications, but because many elderly patients with undiagnosed cognitive impairment undergo surgery and anesthesia, and those patients are at increased risk for postoperative delirium and what amounts to a continuation of the undiagnosed cognitive impairment that was present preoperatively," said Dr. Knopman.

The authors pointed out several factors in the study strengthening the conclusion that there is no association between dementia and exposure to anesthesia. For one thing, it was based on a geographically defined population, thereby reducing the potential for sampling bias at tertiary referral centers.

As well, recall bias was lessened through full access to medical records for a relatively long period before diagnosis (a median of 32 years), strict criteria were applied to diagnose dementia by experienced behavioral neurologists, and the relatively large sample size increased the power to detect associations.

Still an Open Question

Asked to comment on the study, Heather M. Snyder, PhD, director, medical and scientific operations, Alzheimer's Association, said this work hasn't totally closed the book on finding a connection between anesthesia and cognition. Whether there's a link between them "continues to be an open question," she told Medscape Medical News, especially as animal and preclinical work has found an association, she said.

"Some of the animal work has shown that there is accumulation of brain changes associated with Alzheimer's disease when researchers apply anesthesia to the animal models," said Dr. Snyder. "Animal research is so important in helping us understand what's happening in the disease process, but you really need to translate that into human data. This study really moves us forward in our understanding of what the links may or may not be between anesthesia and dementia, but I wouldn't say it definitively answers the question."

Dr. Snyder agreed that POCD is a separate entity. Whether cognitive decline postoperatively increases risk for later dementia also remains "an open question" she said.

The authors did "a nice job" in outlining the limitations of the study, one of which was its retrospective nature, noted Dr. Snyder. "It's looking back in time at medical records, at individuals who had some sort of procedure with general anesthesia and then going forward in time and seeing when and if there was a diagnosis of dementia."

The project was supported by the Department of Anesthesiology, College of Medicine, Mayo Clinic. The study was made possible by grants from the Rochester Epidemiology Project, Mayo Clinic Center for Translational Sciences Activities, and the National Center for Advancing Translational Sciences.

Mayo Clin Proc. Published online May 3, 2013. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: