Neurology -- Correspondence for Carlson et al., 70 (11) 828-833

Correspondence published:

Trajectories of brain loss in aging and the development of cognitive impairment: Owen T. Carmichael, Oscar Lopez, James T. Becker, and Lewis Kuller. (19 June 2008)

Reply from the authors: Nichole E. Carlson, Jeffrey A Kaye (19 June 2008)

Trajectories of brain loss in aging and the development of cognitive impairment 19 June 2008

Owen T. Carmichael,
University of California, Davis
1544 Newton Court, Davis, CA, 95618,
Oscar Lopez, James T. Becker, and Lewis Kuller.
Send Correspondence to journal:
Re: Trajectories of brain loss in aging and the development of cognitive impairment

ocarmichael{at}ucdavis.edu Owen T. Carmichael, et al.

We were pleased to read the report by Carlson et al. on ventricular volume and cognitive decline in an elderly cohort. [1] At a time when the late-life neuroimaging community can be highly “hippocampus-centric,” the study reminds us that ventricular volume is a reliable MRI measure that can be both easy to measure—even in the relatively low-resolution scans used in this study—and predictive of cognitive trajectory.
The editorial by Dr. Weiner mentions that the results "...may not be completely generalizable to other populations." [2] Given the unusual characteristics of this cohort, this point cannot be overemphasized. At enrollment, subjects were in their mid-eighties and free of cardiovascular (CV) risk factors that are highly prevalent late in life and may contribute to dementia in a significant percentage of cases. [3,4] Given that these very common CV risk factors are linked to accelerated ventricular expansion and cognition, it will be important to replicate the study on a more representative cohort to clarify its relevance.
Furthermore, we would be interested to know whether all 79 study participants stayed completely free of all CV risk factors throughout their roughly 2 to 15 year involvement in the study, or how incident CV conditions were dealt with.
Since ventricular volume may be strongly related to survivorship, mortality is another important source of bias to consider in the very old. [5] Given the general consensus that ventricular expansion accelerates later in life, the curious finding that ventricular expansion decelerated with age may simply reflect the deaths of typical subjects, whose ventricles were expanding at an increasing rate until they died. If this is the case, the data may be biased toward representing a relatively unusual group of extremely healthy individuals whose atypical survivorship is reflected in very slow ventricular expansion.
In spite of these possible sample biases, two key results are concordant with those we obtained on a slightly younger epidemiological cohort. [6] In particular, expanded ventricles in cognitively-normal subjects were predictive of subsequent cognitive decline, and ventricular expansion accelerated part of the way through the progression from normal cognition to dementia. The current study has the advantage of more frequent scans, allowing for a more fine-grained analysis of brain change trajectories that accompany cognitive transitions.
Replication of these trajectories in an epidemiological cohort with similarly frequent scans could bolster the utility of serial ventricular volume measurements as a readily available predictor of cognitive function.
References
1. Carlson NE, Moore MM, Dame A et al. Trajectories of brain loss in aging and the development of cognitive impairment. Neurology, 2008;70:828-833.
2. Weiner MW. Expanding ventricles may detect preclinical Alzheimer disease. Neurology 2008;70:824-825.
3. Vasan, R.S., et al., Residual lifetime risk for developing hypertension in middle-aged women and men: The Framingham Heart Study. JAMA, 2002;287:1003-1010.
4. Schneider JA et al., Mixed brain pathologies account for most dementia cases in community-dwelling older persons. Neurology, 2007;69:2197-2204.
5. Kuller, LH, et al., White matter grade and ventricular volume on brain MRI as markers of longevity in the cardiovascular health study. Neurobiol Aging, 2007;28:1307-1315.
6. Carmichael OT et al., Ventricular volume and dementia progression in the Cardiovascular Health Study. Neurobiol Aging, 2007;28:389-397.
Disclosure: The authors report no disclosures.

Reply from the authors 19 June 2008

Nichole E. Carlson,
University of Colorado Denver
4200 E 9th Av., C245 Denver CO, 80262,
Jeffrey A Kaye
Send Correspondence to journal:
Re: Reply from the authors

nichole.carlson{at}uchsc.edu Nichole E. Carlson, et al.

We thank Carmichael et al. for their comments. Our study findings and their implications for new avenues of investigation would certainly benefit from replication, ideally using a variety of study designs. [1] Our study took advantage of a well characterized cohort to advance the understanding of the complex patterns of change in brain volume afforded by annual clinical and MRI assessment. The low burden of vascular risk factors in this group is a strength of this design because conclusions can be drawn regarding the role of brain aging relatively free of vascular disease.
We re-emphasize that the cohort size and design of our study placed limits on the precision of patterns and statistical adjustment that could be made. Accordingly, we focused on understanding the global timing of changes in ventricular expansion relative to cognitive impairment. In the healthy aging cohort described in this study, 49% developed hypertension or cardiovascular disease within 5 years of entry. Those who developed cognitive impairment did not differentially experience stroke or the diagnosis of hypertension or cardiovascular disease compared to those who did not. This finding further supports the notion that these results represent brain aging with less affect of vascular disease.
The mortality rates in those who developed cognitive decline would be expected to be higher since dementia is associated with increased mortality, and thus, the finding of decreasing rates of ventricular expansion is not surprising. We also note that the results in the paper do not suggest ventricular expansion stops at older ages, but rather the rate of expansion was more constant at older ages than at younger ages.
We hypothesized that the deceleration of the accumulation of ventricular rates reflects simple biological and physio-mechanical limits to the expansion of the ventricles with advanced age, but we acknowledge that a survivor effect could be an additional part of this pattern. We are encouraged that our results are further supported by the MRI substudy of the CHS, which found in their predictive models that ventricular size at baseline predicts cognitive outcomes years later. [6]
Our paper offers further expansion on the role of ventricular expansion in MCI by providing a timing to changes, which can only be studied with multiple MRI assessments on a subject. We look forward to other studies addressing this important issue of understanding trajectories of brain volume change associated with aging and age-associated disorders.
Disclosure: The authors report no disclosures.