Our analysis aimed to determine whether F-VEP modifications might be associated to q-EEG changes in AD patients and, in this case, whether the pattern of change were the same in the whole group.
53% of our patients, the AD-AL ones, presented a change in F-VEP occurring selectively under Eye-closed Condition, as previously shown
. The remaining ones, i.e. AD-NL patients, had values within normal limits. In the first case the latency difference between Conditions was significant, in the second one it was not. Question was raised whether the delayed P2 latency might have been a mere artifact. Indeed, Coburn et al
 maintained that, with Eye-open, F-VEP takes the form and the latencies of P-VEP. According to this hypothesis, our study would have compared the latency of P100 in Eye-open Condition with the longer one of P2 in Eye-closed one. If this were the case closure of eyes ought to have been followed by the lengthening of P2 latency in all our groups. Contrary wise, delayed P2 latency occurred only in the AD-AL group. All the others did not differ from normal.
By having ruled out the possibility of an artifact, the increased P2 latency can be considered an indirect sign of functional disorganization of brain activity in Resting State
. This view seems to be confirmed by two concomitant changes of q-EEG spectrum, a parameter apparently independent from F-VEP.
Firstly, the AD-AL group was characterized by A/SW values which, under Eye-closed, were significantly lower than those of Controls due to the collapse of alpha activity and the increase of delta component. The analysis of AD-AL spectral profile confirmed the flattening of the curve where normal peaks were hardly identifiable.
Secondly, AD-AL patients did not present the normal difference between Eye-open and Eye-closed Conditions. In our Controls, indeed, alpha activity, dropped down by about 40% from Eye-closed to Eye-open state, in keeping with other observations
, whereas in AD-AL patients the drop was less than 6%. The opposite occurred for delta activity.
The flattening of the Eye-closed spectrum and the lack of difference between Eye-open and Eye-closed spectral profiles were present in 9 out of 10 AD-AL patients. This indicates that these results are consistent and robust enough to stand out immediately, despite the relative smallness of the group.
In all the AD-NL patients, but three, A/SW ratios were within the normal limits. The mean spectral profile recorded under Eye-closed Condition presented the same fluctuations than that of Controls and differed significantly from the one recorded under Eye-open Condition.
The association of reduced alpha and increased delta activities has been confirmed in AD by a wealth of observations
[13–15]. Lesser attention has been paid to the changes in the spectral pattern of AD when comparing Eye-open Condition with Eye-closed one.
Signorino et al
[16, 17] studied the changes in 6,5-12 Hz band power occurring in AD by analysing the ratio between values observed in Eye-closed and Eye-open Conditions. Such a Reactivity Index was significantly lower than that of Controls, since the alpha activity recorded in AD under Eye-closed state tended to equate that of Eye-open one or to decrease below it.
Other data, derived from non-linear analysis of EEG
, suggest that the levelling off of spectral power in Eye-closed state is not limited to the range of components explored by Signorino et al
[16, 17] but it extends to the whole spectrum of frequencies. Pritchard et al
 observed that in elder Controls the Global Complexity Index obtained under Eye-closed was significantly greater than under Eye-open. The difference faded out in AD patients where the indexes were equal. Overall, the results suggested that the AD process reduces the complexity of cortical dynamics underlining EEG
[13, 36] by affecting the normal capacity to modulate brain activity in response to modified sensory information, such as after closure of eyes.
By showing the lack of difference between Eye-open and Eye-closed spectral profiles, our results confirm Pritchard et al.'s observation
 but specify that changes in Resting State involve only a part, however large, of AD population.
No patient in AD group presented visual signs or symptoms of posterior cortical atrophy
, nor severity of cognitive decline differed between our AD groups. Probably a difference between AD-NL and AD-AL patients, if any, should have been expected in attentional performances. As matter of fact, it is known that impairment in structures related to arousal and alertness significantly correlate with defects of attention
 whose burden does not necessarily parallel severity of cognitive impairment
. The suggestion, however, needs further support as our study did not examine such an aspect.
Since F-VEP and q-EEG changes are strictly limited to Eye-closed Condition, our data support the hypothesis that behavioural states as Eye-open and Eye-closed Rest depend, at least partly, on separate system.
Such a result fits in with recent acquisitions of functional neuroanatomy which point out at different activation patterns between states
[2–4, 33]. In Eye-closed Rest State changing cerebral Rhythms seem to be paralleled by connectivity pattern variations
[40, 42]. Amplitude of EEG alpha rhythm, expression of thalamo-cortical and cortico-cortical synchronization under Eye closed Condition, is associated with changes in fMRI signal in occipital areas and in thalamus
. Correlated activities of medial thalamus and of anterior midbrain seem to precede the start of alpha activity
The functional changes are due, at least partly, to neuronal activity
[3, 42] whose pattern of activation in Eye-closed Condition involves subcortical structures mostly responsible for independent modulation of alpha rhythm and F-VEP.
The impairment of such independent parameters as F-VEP and q-EEG profile under the same behavioural state, indirectly seem to support the network degeneration hypothesis according to which intrinsic connectivity networks might be the selective targets of specific neurodegenerative diseases
[6, 7, 43]. F-VEP and Power Spectrum changes, hence, would witness the involvement of such a network system in our AD-AL patients.
In AD-NL patients, cognitive defects were not associated with changes in the two modalities. Probably the distribution of neurodegenerative process differed from that of AD-AL patients, though both groups presented the same level of dementia. These data suggest that the system damaged in AD-AL group has no effect on cognitive defects at least as they appear from measures drawn from Mental State Scales as MMSE or DRS. At the same time, the occurrence of contrasting results between F-VEP and q-EEG, mainly shown by those AD-NL patients who presented abnormal q-EEG, confirms that the two modalities are functionally independent from each other.
Summing up, F-VEP and q-EEG changes due to AD are likely to be associated to each other under Eye-closed Condition in a part of AD population. Such a group, identified by using F-VEP as a marker, seems to witness a change in the functional architecture of Resting State in Eye-closed Condition. Other patterns of change can occur, as suggested by the heterogeneity of AD population
[18, 44]. Whether these different patterns have distinct anatomo-clinical correlates is matter of further investigation.