Women's superior verbal memory may be an advantage early in life, but perhaps not so much when it comes to early diagnosis of mild cognitive impairment (MCI).
New research, published online March 16 in Neurology, reveals that despite similar levels of hippocampal atrophy, older women have better verbal memory than older men.
Because verbal memory deficits are used to diagnose amnestic MCI (aMCI), women could be missing out on early detection of this condition, said study author Erin Sundermann, PhD, postdoctoral research fellow, Einstein Aging Study, Albert Einstein College of Medicine, Bronx, New York.
Norms for tests to determine aMCI and Alzheimer's disease (AD) are adjusted for age and education, but not for sex, she said.
The new study included 1308 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) who had normal cognition (controls) or had been diagnosed with aMCI or AD dementia. As well as diagnostic information on these participants, researchers had data on hippocampal volume and ratio of hippocampal volume to intracranial volume (HpVR).
They also had patient scores on verbal episodic memory from the Ray Auditory Verbal Learning Test (RAVLT). The range for RAVLT immediate recall is 0 to 75, and the range for RAVLT delayed recall is 0 to 15.
The researchers looked at the overall sample to get an idea of the relationship between clinical markers, such as verbal memory, and pathophysiologic biomarkers, such as hippocampal atrophy.
They found that women and men in the study had similar hippocampal atrophy as measured by HpVR. But the mean score for RAVLT immediate recall was 38.1 for women and 32.7 for men, and scores for RAVLT delayed recall were 5.1 and 3.8, respectively (P < .001 for both).
The researchers also found a significant sex-by-HpVR interaction for both immediate (P = .001) and delayed (P = .008) recall on the RAVLT in the overall sample. The magnitude of this positive association was stronger in women than men for immediate and delayed recall.
"We're finding that women have better verbal memory at the same level of evidence of hippocampal atrophy," said Dr Sundermann.
The sex difference uncovered by the study might reflect a form of cognitive reserve in women that delays verbal memory decline until the disease becomes more advanced. According to the theory behind cognitive reserve, factors such as education, higher IQ, and occupational achievement allow patients to compensate for neuropathologic changes by, for example, engaging alternative brain networks.
But while the HpVR-by-sex interaction was significant in the study's aMCI group, it wasn't significant in the AD dementia group.
Dr Sundermann and her colleagues had predicted that the sex differences "would be completely abolished" in the AD group. However, while it was abolished for delayed recall, women maintained an advantage over men on the RAVLT immediate recall score, although Dr Sundermann noted that "it was much, much smaller" in the AD group compared with controls and the MCI group.
"Maybe there's not a complete abolishment of the sex difference, but it's highly minimized in the AD stages," she said.
The study didn't find a relationship between hippocampal volume and memory performance in healthy older adults. Previous ADNI studies had shown similar findings, so this wasn't unexpected, said Dr Sundermann.
"We think this lack of relationship may reflect a threshold effect," she said. "As long as the hippocampal volume remains above a certain level, it may not be associated with memory."
The results might help explain what the authors called the "paradoxical" sex differences in the incidence of aMCI vs AD dementia.
Some studies have reported that more men are diagnosed with MCI whereas more women have AD dementia. In the current study, a greater proportion of men than women had aMCI (56.8% vs 48.5%; P < .002) and a greater proportion of women had AD, although this wasn't significant (19.7% vs 17.8%; P = .08).
"What may be going on is that if women have this advantage in verbal memory, they are perhaps getting diagnosed with MCI later than men, and then once their decline starts, they're declining faster," said Dr Sundermann.
It's possible that women who have MCI during this "shorter window of time" are being missed in studies that assess patients maybe only once every year or two, said Dr Sundermann.
She noted that verbal memory is typically the first cognitive domain that declines in AD, and it's tests of this domain that are used to diagnose MCI and AD.
Dr Sundermann and her colleagues now plan to calculate norms based on sex, as well as on age and education, for these cognitive tests.
"We will then apply those to the data and see if perhaps we're picking up more women with MCI who with previous norms were identified as normal because their advantage in verbal memory might have masked any pathology that they already had."
Mary Sano, PhD, director, Alzheimer Disease Research Center, professor, Department of Psychiatry, Mount Sinai School of Medicine, and director of research, James J Peters Veterans Affairs Medical Center, Bronx, New York, coauthored an editorial that accompanied the study.
A "take-away" from this study for clinicians is that they might consider baseline cognitive testing in women, Dr Sano told Medscape Medical News. "The most important thing is not what your score is, but whether it's a change from where you used to be."
Clinicians might also consider taking more seriously a female patient's subjective memory symptoms, even if their initial impression is that the patient is cognitively intact, said Dr Sano. "Women's cognitive resilience is so high that unless you have an opportunity to watch whether or not it's changing, you might miss it."
The field in general needs to find better ways to detect early cognitive changes in women, possibly with more challenging tests or with better normative data, said Dr Sano.
Dr Sundermann and Dr Sano have disclosed no relevant financial relationships.