Body Mass Index Trajectories Preceding Incident Mild Cognitive Impairment and Dementia

From the JAMA Network, this is JAMA Psychiatry Author Interviews, conversations with authors exploring the latest clinical research, reviews, and opinions featured in JAMA Psychiatry.

Dr Torous: Welcome to our JAMA Psychiatry Author Interview podcast. I'm your host and web editor, Dr John Torous. Today we're talking about a brand new paper just out in JAMA Psychiatry about body mass index trajectories preceding incidents of mild cognitive impairment and dementia in the role of brain pathology. We're about to be joined by one of the two senior authors from the paper, Dr Xu, PhD, from the Aging Research Center in the Department of Neurobiology, Care Sciences and Society at the Karolinska Institute in Stockholm, Sweden.

We're all aware that physical health and brain health are linked, with the body and brain connection certainly well recognized. But what exactly is that link? A better understanding of this connection could help us realize mechanisms of brain illnesses and move the field towards prevention strategies. Thus, this new paper just out in JAMA Psychiatry on the relationship between body mass index and mild cognitive impairment in dementia is certainly very timely. So, thank you very much for joining us, Dr Xu, and could you start by telling us what is mild-cognitive impairment, and how is it related to dementia?

Dr Xu: So mild-cognitive impairment, MCI, is a prodromal phase of dementia and the transition state between normal cognition and dementia. MCI affects about 16% to 20% older adults, and more than one third of MCI cases go on to develop dementia over five or more years. But not all people with this MCI go on to develop dementia, so MCI is regarded as a re-workable state before dementia. And also, MCI provides an opportunity to prevent or delay dementia onset. Therefore, identifying a risk factor for MCI may help us to slow down or prevent the development of dementia.

Dr Torous: That's very helpful. An- and your paper focuses on body mass index, BMI. Can you help us understand what is BMI, and what was known about its relationship to brain health before your study?

Dr Xu: BMI is calculated as weight in kilogram divided by square height in meter. In this study, weight and height were measured at baseline and during annual followups by a trained technician of participants with light clothes and no shoes. [So] we could have the opportunity to trace long-term trajectory of BMI. Regarding the BMI in relation to brain health, a few studies using brain-pathological data have shown an association between higher burden of AD pathologies and a lower average BMI before death. But they found no significant association between BMI and the cerebral-vascular pathologies. So rather than using average BMI across the followups to evaluate the BMI change, they used the mixed-effect model analysis with a backward time scale, accounting for variability of BMI over time to trace the trajectory [until] MCI or dementia onset, so we can have the opportunity to see the long trajectory during the long follow-up.

Dr Torous: Got it. That's very helpful. And in your study, you also cover brain pathology. Certainly, this is a psychiatry podcast, so can you give listeners a background on what part of brain pathology is relevant to understand your paper and results?

Dr Xu: Yeah. Regarding pathology, there is really limited evidence about the association between brain pathology and later-life BMI. Mechanisms underlying the class in the pre-clinical and prodromal phase of dementia are not well understood. We found that BMI declined more steeply in those with a high versus low burden of AD pathology or cerebral-muscular disease among dementia-free participants. Therefore, our findings suggested that higher level of AD pathology or vascular disease may underlie the BMI decline proceeding the MCI. And also this finding suggests that monitoring weight change may help to identify higher-risk population in the early stage of MCI for early prevention, to prevent or delay dementia onset.

Dr Torous: And in building up to those results in that, you drew your population from the Rush Memory and Aging Project.

Dr Xu: Yes.

Dr Torous: Can you tell us more about what is this project, and why is it a useful resource for doing this type of work?

Dr Xu: The Rush Memory and Aging Project, we called MAP, this is an ongoing, longitudinal cohort study, investigating risk factors for common, chronic, neuro-degenerative conditions in older adults. And the study participants were recruited from continuous care, retirement communities, senior and subsidized housing, church group, social service agencies, and the individual home in northeastern Illinois and the Chicago area.

So between 1997 and 2020, MAP enrolled 2,192 participants who were annually followed up through 22 years. So this is a long follow-up period. And during the followup, a total of 520 participants died and underwent a brain pathology. So this community-based cohort study with long term, yearly follow-up gave us the possibility to trace the BMI trajectory and their onset and also their relation to brain pathology.

Dr Torous: And from this cohort of people, what were the inclusion/exclusion for people that were included in your study, of course, focusing on BMI and MCI and dementia?

Dr Xu: Yeah. In our study, of this more than 2,000 participants, we included only people who were cognitively intact, after excluding people with prevalent MCI or prevalent dementia as well as those people who had missing information on BMI.

Dr Torous: And for these people that, again, met your inclusion criteria, can you review what data or measures were you pulling from them, or kind of measuring from them or kind of drawing from the Rush, Memory, Aging study? What were the measurements on them?

Dr Xu: Yeah. We have, like, comprehensive data in this project that were collected. For example, information on social demographic factors, such as age, sex, education, lifestyle factors, chronic disease, including diabetes, vascular disease, and also that we have also genetic data collected, for example with APOE ɛ4 genes also collected, and then measured from the blood sample. And all participants underwent a comprehensive, clinical evaluation, neurological examination and extensive cognitive tests at enrollment and followup visits thereafter. So during the 22 years, we followed them each year, this is very valuable in the data collection part.

Dr Torous: Yeah, that's certainly a- a lot of data. So-

Dr Xu: Yeah.

Dr Torous: ... let's pause again to review your main hypothesis around, again, the connection between BMI, body mass index, MCI, mild-cognitive impairment, and dementia. What was the main hypothesis, again, you were looking for, now that we understand a little bit about these concepts and the data that you're working with?

Dr Xu: Yeah. When we started this study, based all the current literature, the main hypothesis of our study was that BMI may decline many years before MCI onset, and that the accumulation of both Alzheimers and the vascular pathologies in the brain might underlie this BMI decline leading up to MCI and subsequent dementia.

Dr Torous: And can you tell us about the final cohort that was included from the Rush project and what were the characteristics of the people that you included in your analysis?

Dr Xu: The participants in MAP were volunteers from the community, and they had higher level of education and performed well on cognitive tests. However, the characteristics of the MAP participants are generally similar to those in other well-established cohorts, such as the Honolulu-Asia Aging Study and The Kungsholmen Project in Sweden, in terms of demographic factors and age at MCI and dementia onset. However, caution is needed when generalizing our findings to other populations, especially to young or old adults.

Dr Torous: And let's talk about now the main result, and I'll break this question up into three parts. First, what were the main results about BMI, or body mass index, and mild cognitive impairment?

Dr Xu: What we found in this study among people who were cognitively intact, the BMI tended to decline earlier and faster in people who developed MCI compared to those who did not. And a significant BMI decline occurred beginning around seven years before MCI onset.

Dr Torous: Interesting. And how about for dementia, were the results the same or different?

Dr Xu: Yeah, in the dementia risk we evaluated in the MCI cohort, the results indicated that after MCI diagnosis, BMI declined at the same pace in people who developed dementia and those who did not. So during the MCI development, BMI declined faster and steeper, but after MCI, the BMI declined parallelly among people who developed dementia and who did not.

Dr Torous: Those results, were they similar or different for brain pathology?

Dr Xu: Actually, the brain-pathology cog-... data just provided us further explanation or as mechanisms to support what we observed for MCI and dementia. In the pathological data analysis, we found that BMI declined faster among participants with a higher burden of global AD pathology or vascular pathology. Therefore, AD and vascular pathologies might explain the BMI decline during the development of MCI. In other word, BMI decline might be an indicator for the presence of brain pathologies.

Dr Torous: Which is a very novel and interesting result.

Dr Xu: Yeah. This is the first study we found that report this.

Dr Torous: I know we talked in the beginning about some of the prior studies. How did your results fit into, kind of, the prior literature, or is it just very different, these results?

Dr Xu: As it showed from many other studies, there is a very directional association between BMI and the dementia risk. That means higher BMI middle life... High BMI, it means that BMI greater than 25, including overweight and obesity. But later life, low BMI has been associated to both MCI and dementia risk, although with some inconsistent results. Regarding BMI changes, several studies have explored BMI change that occur in the diagnosis data of dementia, reporting that BMI decreased faster during the clinical phase of dementia than in healthy aging, thus, conferring a lower BMI at the time of dementia diagnosis. However, evidence on the BMI trajectories before MCI diagnosis is unknown. In our study, targeting these knowledge gaps, we conducted this study to address questions about when and to what extent the BMI declined prior to diagnosis of MCI and subsequent dementia, and to explore whether AD and vascular brain pathologies may explain such decline.

Dr Torous: And certainly, we know that correlation is not causation. But your results are certainly very exciting. How would you imagine that clinicians could begin to use this today?

Dr Xu: Yeah. That's a really good question. And the findings from our study highlight the importance of monitoring weight change regularly among older adults, as a faster declining BMI and a lower BMI may indicate the need for cognitive assessment. And monitoring weight change may help to identify high-risk population in the early stage of MCI.

Dr Torous: Excellent. So I think this gives us a lot to think about, a lot to even begin to have discussions with our patients. Of course, there are so many more details in the excellent paper that you and the many coauthors have done. So I want to thank you again, Dr Xu, and thank you to our listeners for tuning in. This episode was produced by Shelly Steffens at JAMA Network. The audio team here also includes Daniel Morrow, Jessie McQuarters, Lisa Hardin, and Audrey Forman.

To follow this and other JAMA Network podcasts, please visit us online at jamanetworkaudio.com. Thanks for listening.