Encephalitis lethargica

The term "bradyphrenia' was introduced in 1922 by Naville [1] to describe the slowing of cognitive processing associated with parkinsonism, following pandemic encephalitis lethargica. It was almost pathognomonic of this condition, but had been noted previously in classical Parkinson's disease. Wilson in 1947 [2], further described this 'lethargy of the mind' as 'distinguished by a lack of interest, initiative, attention, concentration: by fatigue and slow reactions: uncommunicative and wishing to be left alone, the patient sits and does nothing unless exhorted to rouse himself'. However, 40 years later, there was still controversy [3-5] over whether the cognitive impairmented existed as a separate entity in idiopathic parkinsonism, or could be explained by dementia, age, and/or co-existent depression, or was iatrogenic [6, 7]. As so often happens in medicine, the argument seems to have devolved into a measure of the 'standing' of the schools involved. Some even found semantic objections to the term 'bradyphrenia' sufficient to deny the existence of the cognitive impairment which it was coined to describe. Alternative terms, such as 'psychic akinesia' and 'subcortical dementia' are also questionable, semantically and pathophysiologically; we favour the original coinage.

The lack of a simple model against which to test for the presence or absence of bradyphrenia was a major problem. In the 1980s, there was a flurry of papers measuring reaction time in Parkinson's disease, but this was an inadequate model for bradyphrenia, since it has a motor as well as a cognitive component. Bloxham et al. [8] hit upon the elegant notion of using the difference between reaction times to an unpredictable and a predictable imperative signal: the amount of quickening in response to a warning is, surely, a reflection of the cognitive efficiency in processing that warning. The imperative signal used was 'GO LEFT' or 'GO RIGH', the subject being expected to raise the corresponding index finger from its touch sensitive support. Prior to this was an alerting signal of 'READY', 'READY LEFT' or 'READY RIGHT', so that the subject either did not or did know in advance which side was to be lifted. The investigators [8] found that sufferers appeared less able to use predictability to reduce reaction time, but studied only nine parkinsonians and nine age-matched controls. We have been able recently to confirm their findings, studying 103 subjects with idiopathic parkinsonism and 144 without [9].

A quickening in response to pertinent information is not necessarily related inversely to disease-induced slowing of information processing. A more direct measure of the latter would be a slowing, with increasing cognitive complexity, in parkinsonians, beyond that shown by a control group [5]. Delayed auditory feedback (DAF) is a method of increasing the cognitive complexity of speaking exercises. It extends the time between utterance of speech and its auditory perception. This slows speech and induces errors. Moreover, DAF, as an objective test for the presence or absence of bradyphrenia, would have the advantage over reaction time testing that it does not require the exercise of manual skills. Delayed auditory feedback did, indeed, disrupt the speech of 104 subjects with idiopathic parkinsonism more than that of 144 healthy controls [10]. Three types of exercise were employed, counting, repeating standard sentences in turn after the prompter and reading words from a card. Three different test conditions were used, two delays in feedback, 1/6s and 1/3s, being compared with no delay. A 1/6s delay, during the more spontaneous exercise of counting, discriminated best between those with and without parkinsonism, the increment in performance time over that with no delay being unrelated to the intrinsic speed of vocationalization.

Cognitive disability in parkinsonism needs to be subdivided according to the influence of general brain damage, depression and putative specific deficits, such as those in visuospatial function, recall memory, sequencing, temporal ordering, recency discrimination, spontaneous generation of tasks, specific planning [5] and bradyphrenia. Could bradyphrenia be just a part of a dementia syndrome [11]? Moderate to severe functional dimentia on a global scale was a ground for exclusion from our studies on the efficiency of response in reaction time to a warning and the disruptive effect on speech of DAF [9, 10]. Moreover, the global cognitive function score explained only a small (11%), albeit statistically significant, amount of the variance in reaction time response to a warning (and that only in those with parkinsonism), and appeared irrelevant to the disruptive effect of DAF.

Are the efficiency of the response to a warning and the disruptive effect of DAF congruent or complementary is discriminating between those with and without parkinsonism? In out work [9, 10], they were found to be complementary, irrespective of the order in which they were added to a model already containing the global cognitive function score. None of the cognitive deficits mentioned is considered to be a cardinal sign of parkinsonism. Howerver, the discrimination between those with and without clinical parkinsonism by means of a simple global cognitive function score and two measures of bradyphrenia was remarkable. By calculation of the cross-validated, predicted probability of being parkinsonian for each of 247 subjects, and selection of a cut-point of 0.5 on the predicted probability of being parkinsonian, clinical parkinsonism could be predicted with a sensitivity of 42% and a specificity of 86%.

What are the influences of other subjects characteristics on bradyphrenia? If there is an age-related tendency towards parkinsoniasm [12], one would expect any cognitive disability associated with the idiopathic syndrome to reflect this. A small, but significant, effect of ageing was seen both on the efficiency of response tgo a warning and the disruptive effect of DAF in those without clinical parkinsonism: early evidence of bradyphrenia was a feature of ageing.

Depression is common in sufferers from parkinsonism, but it influenced psychomotor performance, as measured by total reaction time, and not the efficiency of response to a warning [19]. Thus, the mechanisms by which depression and parkinsonism prolong reaction time are not the same [4]. Depression did not influence the disruptive effect of DAF [10].

We conclude that the changes in central processing time in idiopathic parkinsonism had a component which could not be explained by a glibal cognitive function score, the ageing process or depression. Bradyphrenia appears to be not only a nosological entity, but also a feature of an age-related tendency towards parkinsonism.

Idiopathic parkinsonian is a diverse condition. Is, then, the cognitive disability linked with any of the cardinal signs? Interestingly, the disruptive effect of DAF was greater in those with less prominent, or no, tremor [10]. This might be explained by the admixture of aetiological influences, an environmental influence being perhaps more important in the development of a rigid, akinetic [13], bradyphrenic syndromde, and a genetic influence more important in a predominantly tremulous sundrome [14], with little bradyphrenia.

Could the cognitive impairment be iatrogenic? The presence of bradyphrenia could not be explained by sedative or alcohol consumption. Does the relatively low sensitivity, but high specificity, of the measures of bradyphrenia, in discriminating between those with and those without clinical parkinsonism, reflect some reversal of the parkinsonian deficit by exogenous substances? Just as parkinsonism is less common in tobacco smokers, the bradyphrenia of parkinsonism, as measured by the efficiency of response to a warning, appeared less pronounced in smokers, despite withdrawal over the morning or afternoon of the experiment. Anti-parkinsonian medication appeared to facilitate the response to a warning. Almost all treated sufferers were receiving maintenance levodopa therapy, alone or in combination. The disruptive effect of DAF in parkinsonism was not modulated by the presence of these exogenous substances. The way is now open to precise psychometric mapping of the probably complex pattern of drug responsiveness, using prospective placebo-controlled protocols.

The delay in confirming and quantifying this potentially treatable aspect of parkinsonism results not from the dogmatic expression of opinion, but from a human readiness to follow conventional thinking. This provides a trouble-free existence for physician and research worker, but a poor example to their students. The potential availability of data has never been a problem in geriatric practice; it is in defining the key questions, and obtaining adequate statistical advice, that the difficulty, and the excitement lie. After all, Geriatric Medicine owes its existence to individuals who dared to think.

References

[1.] Naville F. Etudes sur les complications et les sequelles mentales de l'encephalite epidemique. Encepale 1922;17:369-75.

[2.] Wilson SAK, Neurology. London: Arnold, 1947.

[3.] Mayeux R, Stern Y, Sano M, et al. Clinical and biochemical correlation of bradyphrenia in Parkinson's disease. Neurology 1987;37:1130-4.

[4.] Rogers D, Lees AJ, Smith E, et al. Bradyphrenia in Parkinson's disease and psychomotor retardation in depressive illness. Brain 1987;110:761-76.

[5.] Brown RG, Marsden CD. Cognitive function in Parkinson's disease from description to theory. Trends Neurosci 1990;13:21-9.

[6.] Riklan M, Whelihan W, Cullinan T. Levodopa and psychometric test performance in Parkinsonism--5 years later. Neurology 1976;26:173-9.

[7.] Agid Y, Ruberg M, Dubois B, et al. Biochemical substrates of mental disturbances in Parkinson's disease. In: Haasler RG, Christ JF, eds. Advances in neurology, vol. 40. New York: Raven Press, 1984;211-18.

[8.] Bloxham CA, Mindel TA, Frith CD. Initiation and executive of predictable and unpredictable movements in Parkinson's disease. Brain 1984;107:371-84.

[9.] Dobbs RJ, Bowes SG, Charlett A, et al. Hypothesis: the bradyphrenia of parkinsonism is a nosological entity. Acta Neurol Scand 193;87:255-61.

[10.] Dobbs RJ, Bowes SG, Henely M, et al. Assessment of the bradyphrenia of parkinsonism: a novel use of delayed auditory feedback. Acta Neurol Scand 1993;87:262-7.

[11.] Revonsuo A, Portin R, Koivikko L, Rinne JO, Rinne UK. Slowing of information processing on Parkinson's disease. Brain Cognition 1993;21:87-110.

[12.] Dobbs RJ, Lubel DD, Charlett A, et al. Hypothesis: Age associated changes in gait represent, in part, a tendency towards parkinsonism. Age Ageing 1992;21:221-5.

[13.] Kirollos C, O'Neill CJA, Dobbs RJ, et al. Quantification of the cardinal signs of parkinsonism and of associated disability in spoues of sufferers. Age Ageing 1993;22:20-6.

[14.] Barbeau A, Roy M. Familial subsets in idiopathic Parkinson's disease. Can J Neorol Sci 1984;11 (no. 1, suppl.):144-50.

COPYRIGHT 1994 Oxford University Press
COPYRIGHT 2004 Gale Group