2003, Number 4
Salud Mental 2003; 26 (4)
Deterioro cognoscitivo incipiente: un estudio con SPECT de activación y neuropsicología
Castañeda M, Ostrosky-Solís F, García RJC, García MJA, Gutiérrez LM, Medina V, Heinze MG
PDF size: 413.96 Kb.
ABSTRACTMild cognitive impairment (MCI) is a borderline condition between normal aging and dementia and is characterized by subjective complains of memory impairments that go beyond those expected considering age and education. Yet subjects are not demented, and their functional status remains intact (Peterson et al. 1999).
It is suspected that this group includes a substantial number of patients with pre-clinical Alzheimer’s Disease(AD), since the follow-up of this group indicates a conversion rate from MCI to AD of 12 to 25% per year and 50% convert in 5 years. Normal controls by contrast, convert to AD on an average of 1-5% per year, depending on age. The identification of people at potential risk of dementia, could be helped by an early therapeutic intervention and also, it may lessen distress for both patient and family, minimize the risk of accidents, and perhaps even prevent the onset of the dementig process itself.
Neuropsychological assessment is relevant to make a differential diagnosis between normal and pathological aging, to distinguish between different types of dementia, to define the patterns of strength and weakness and to suggest the likely pattern of underlying cerebral pathology. The Single Photon Emission Computerized Tomography (SPECT) is one of the recent neuroimaging techniques that contribute with information about the regional blood flow. Several studies of regional blood flow in AD patients demonstrate relative temporal and parietal hipoperfusion in AD. SPECT imaging conducted while the patient is engaged in a cognitive task or under sensory stimulation are referred to as activation studies. Activation studies afford unique opportunities to explore brain metabolic changes related to specific cognitive operations and to establish hypothesis of the neural networks supporting very discrete cognitive functions. Recent functional neuroimaging studies used during cognitive tasks have added to our understanding of the neural anatomy of cognition in both normal and pathological states; therefore the application of this technique to the study of patients with mild cognitive impairment could provide additional information for the early identification of this disease. Although activation studies have been used with EA, there are very few studies that have used activation methods to study MCI. Using SPECT, Riddle et al. (1993) studied ten patients with EA and nine age-matched normal controls with a verbal memory activation task and found significant differences between the groups only during the activation task.
Since the differential diagnosis between normal aging, MCI and depression is still a matter of controversy, activation studies could provide objective data for the early and objective diagnosis of this group. We performed SPECT perfusion imaging during a basal and during a recognition verbal memory task in a group of normal and MCI subjects. Twenty-three subjects were studied ten controls and thirteen with MCI, matched by age and education. Patients and controls were not receiving psychotropic drugs. The Clinical Memory Unit and the Geriatric Service, of the Instituto Nacional de Nutrición Salvador Zubiran of Mexico City referred the subjects.
The diagnosis of MCI was made if the patients met the following criteria: 1) memory complaint confirmed by a relative, 2) normal activities of daily living, 3) normal general cognitive function confirmed by Neuropsychological Testing, with scores within 1.00 to 1.5 deviation below the norms on the Brief Neuropsychological Test for Spanish Speaking Subjects (NEUROPSI) and on the NEUROPSI Attention and Memory Test, 4) not demented according to the Diagnostic and Statistical Manual of Mental Disorders, Revised Fourth Edition, and the National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer’s Disease and Related Disorders of Association criteria respectively. Control subjects were chosen from the community population of individuals receiving general medical examination or among those who were related to patients. Patients and controls gave written consent.
Neuropsychological assessment was carried out one week before the scanning. SPECT studies were carried out in the Department of Nuclear Medicine of The National Institute of Psychiatry with a Single Slice, multi-detector, three-head scanner (Siemmens) with 70 multiplexors per detector. Tc99m-ECD was used as tracer. Following the paradigm proposed by Riddle et al. (1993) a verbal memory task was implemented. Stimuli included:
a) list of stimuli with “yes” “no”, previously recorded by a feminine voice, with a rate of presentation of 1 each 5 seconds,
b) a list of 10 high frequency words for Spanish language was presented for as many trials as required for five out of ten words to be recalled. The recognition component of the test was performed 5 minutes after,
c) a recognition list which comprised 50 words "10 target" words repeated twice and at random, and 30 semantically and phonetically similar distractor words. Subjects were asked to respond “yes” or “no” according to whether they recognized the word. The recognition score was calculated by awarding +1 for a correct recognition, -1 for an incorrect recognition and 0 for a failed recognition. Maximum score was +20, the minimum –30.
Each of the subjects was scanned on the same morning using a split-dose of Tc99m-ECD tracer. An in-dwelling intravenous catheter was inserted in an arm vein 15-20 minutes before the first injection of tracer (250 MBq). During injection and for five minutes afterwards, patients were reclining with eyes covered with patches and ears unplugged; background noise was minimal.
For the baseline condition, the first injection of tracer was given over 30 seconds while subject repeated “yes” or “no” after the voice previously recorded, at a rate of 1 per second over 5 minutes beginning 1 minute before the injection. The subject was then scanned.
For the activation condition, before the second injection of a tracer (again 250MBq) all subjects were given up to 5 presentations of the 10 word list, until 5 words had been recalled. Beginning 15 seconds before the second injection of tracer, recognition of the 10 words from the 50 word recognition list was tested over 4 minutes at a rate of 1 word per 5 seconds. Subjects again responded “yes” (for a target word) or “no”(for a non-target word), so that simple verbal output was identical between the comparison conditions. Duration of each condition was approximately 15 minutes. State anxiety levels were assessed by administration of the Alderly Park State Anxiety Questionnaire (Walker, 1990) 5 minutes after each injection.
The regions of interest that were analyzed included: In the lower cut: anterior cingulate, frontal, superior temporal, midtemporal, posterior cingulate and occipital. In the upper cut: Anterior cingulate, frontal parietal, posterior cingulate and occipital, and the subcortical regions analyzed were: caudate, putamen and thalamus. The cerebellum was chosen for normalization. In each region pixel counts were performed.
Activation indexes were calculated according to the following formula: Basal R=basal count/ basal cerebellum. Activation R=activation count-basal activation/basal cerebellum. Activation Index =(Activation R-Basal R)/(Activation R+ Basal R). The significance of changes in tracer uptake between baseline and activation conditions for individual regions of interest within each group was assessed using paired Student's t –tests. With the index obtain with the formula, a Student-T test for independent samples was used to compare MCI and control subjects.
During the SPECT studies three cases with multinfarts were detected, and therefore were eliminated from the MCI sample. No significant differences were found in the learning of the word list, the control group had an average of 9.2 (1.2) out of 10 words, and the MCI obtained an average of 9.0 (1.0) out of 10 words. The control group achieved a recall of five words within five presentations and the MCI group within the seventh presentation. Nevertheless no significant differences were found in the recognition list. Performance list was more variable and with higher rate of errors in the MCI group than controls.
Control subjects showed significant increases in uptake during the recognition tasks in several regions, including: In the lower cut: right anterior cingulate, right and left superior temporal, mid-temporal, left posterior cingulate. In the upper cut: right parietal, right posterior cingulate, right and left putamen, right thalamus. The MCI subjects showed differences in several subcortical structures including the right and left thalamus and left caudate nucleus in the lower cut.
A great deal of interest has been generated concerning the topic of a boundary or transitional state between normal aging and dementia of the Alzheimer type. This condition has received several descriptors including mild cognitive impairment, incipient dementia, and isolated memory impairment. Although the criteria for MCI have been accepted, the operalization of these criteria can be challenging. In the present study, although no significant differences were found in the total learning of the word list, and differences were related to the rate of learning, the control group achieved a recall of five words within five presentations and the MCI group within the seventh presentation. During the SPECT activation task, significant differences between groups were found specifically in subcortical regions. Regional blood flow is tightly coupled to local neural demands for glucose. Changes in the pattern of regional cerebral perfusion thus provide an index of cerebral metabolic activity. At baseline, uptake tracer was almost identical in MCI subjects and controls whereas during the activation task the control subjects showed significant activation effects while MCI showed hipoperfusion of several subcortical structures including the left thalamus and caudate nucleus. Similar findings were reported by Riddle et al. (1993) who found bilateral activation of the frontal cortex, anterior cingulate and left parietal in normal controls but not in patients with AD during the same recognition memory task. In previous studies of regional glucose metabolism with Positron Emission Tomography (PET), subcortical abnormalities appear to precede cognitive impairment. It suggests that a gradual phase of neuronal degeneration can be detected and preced clinical diagnosis by many years. It has been reported that there exists a period of several years between detectable disease and clinical diagnosis; this offers the possibility of therapeutic intervention at a stage when most cognitive functions are still preserved. Therefore it opens the possibility for disease onset to be postponed or perhaps even avoided. The dividing line between normal physiological evolution and disease will continue to be difficult to define, and although this is a transversal study with a very small sample, results are encouraging since by performing it together with the neuropsychological testing SPECT studies, it might help to identify preclinical stages of dementia.