Cranial computed tomography of elderly patients: an evaluation of its use in acute neurological presentations.
Computed tomography (CT) is a widely available investigation for patients presenting with acute neurological dysfunction. However, few studies have specifically evaluated CT brain scans in the management of elderly patients presenting in this way. The Oxfordshire Community Stroke Project  suggested criteria for the urgent CT scanning of patients with |stroke' [2, 3] and these were: (a) doubt--either due to inadequate history or a suspected treatable intracranial lesion; (b) suspected cerebellar haemorrhage or infarction; (c) the exclusion of intracranial haemorrhage in patients taking or likely to need antihaemostatic medication; and (d) a patient deteriorating in a manner atypical of a stroke. The population in that study comprised all age groups. Elderly patients with acute neurological syndromes may be difficult to assess, as a complete history of circumstances prior to admission can be difficult to ascertain and subsequent management, particularly surgical intervention, is likely to differ from that in younger patients. We examined the application of these criteria for cranial CT in elderly patients and assessed whether the information obtained influenced subsequent patient management. An attempt is made to suggest particular criteria for scanning in order to yield more positive results in this group.
Patients and Methods
A retrospective evaluation was made of 100 consecutive cranial CT scans of patients admitted under the care of the Geriatric Medicine departments at St George's and the Bolingbroke Hospitals, London. The clinical records from these patients were reviewed.
Patients were grouped according to their presentations and fell into the following categories:
1. Acute confusion, defined as recent loss of cognitive function with a lowered standard abbreviated mental test score.
2. History of falls or other aspects of history such as head injury or fluctuating conscious level to suggest possible subdural haematoma.
3. Patients presenting with transient ischaemic attacks/transient neurological deficit.
4. Patients receiving anticoagulant therapy and suspected of having intracerebral haemorrhage.
5. Suspected cerebellar haemorrhage or infarction.
6. Progression of neurological signs and deterioration following stroke.
7. Atypical neurological signs in a patient presenting with |stroke', or suspected space-occupying lesion.
The CT report findings were assessed as |major', |minor' or |negative'. A major finding was defined as a treatable lesion or a result that directly altered patient management. Minor findings were abnormal scans showing pathology but which did not lead to an alteration in treatment. A negative scan was defined as one which yielded no new diagnostic information and therefore did not alter patient management. Previous studies have failed to correlate cerebral atrophy with impaired cognitive function [4, 5]. In our study we regarded the finding of cerebral atrophy as not providing diagnostic information.
Of the 100 consecutive scans reviewed, the notes of 93 patients were obtained, the remaining seven were not traced. The demographic details of the patients studied are shown in the Table and the effects on outcome in the Figure.
Table. Age and sex of 93 patients studied Age (years) Women Men Total 70-74 7 6 13 75-79 15 12 27 80-84 14 11 25 85-89 18 5 23 >90 2 3 5 Mean age (SD) = 80.4 (5.8) years.
All of the seven patients found to have treatable lesions in the group with confusion were successfully treated. Four had subdural haematomas, one a midline oligodendroglioma and one a cerebellar haemorrhage. In this group, 11 of the 21 patients had scans producing diagnostic information (52%, 95% confidence interval: 31-73%). Only one of the 23 patients with a history specifically suggesting the possibility of subdural haematoma had a positive scan and underwent surgery successfully. With regard to patients referred for anticoagulant therapy, none of the patients presenting with clinical transient ischaemic attacks had abnormal scans. Three patients with minor stroke and scanned prior to anticoagulant therapy had scans resulting in an alteration of treatment. Eight patients, altogether, were scanned for suspected cerebellar lesions; three patients had cerebellar infarcts. Six patients were referred because of progression of stroke and the results did not affect outcome. Five of these scans confirmed large cerebral infarcts and four patients subsequently died. Of the fourteen patients referred because of atypical signs of stroke, seven were found to have treatable lesions. Twelve of the patients with atypical signs of a stroke had scans yielding positive diagnostic information [86%, confidence interval (CI): 76-95%].
In all, 19 (20% CI: 12-28%) of scans resulted in a major outcome as defined earlier of which nine resulted in successful treatment. The greatest proportion of positive scans (13) for treatable lesions was found in the groups presenting with confusion and in those presenting with signs atypical of stroke (68%, CI: 48-89%). Twenty-nine (31% CI: 22 41%) were positive for pathology of minor diagnostic importance. Thus, in our series of 93 patients, CT scan contributed to the management in 52% (CI: 46 62%).
Patients who are found to have subdural haematomas can benefit dramatically from surgical intervention ; but since relatively minor injuries can cause such a lesion, they may be easily overlooked or forgotten . Suspecting the diagnosis can be difficult as there is no characteristic presentation for subdural haematomas in elderly people. In our study, it was not the group with a clear history of falls and neurological deterioration that yielded the highest incidence of subdural haematomas, but the group presenting with confusion. None of these patients had focal neurological signs but presented with hallucinations and personality change of recent and dramatic onset.
Our results from scanning patients with confusion seem to correlate with the findings of Roberts and Caird  who found that of 170 elderly patients with confusion of less than 1 year, 31% had potentially treatable lesions.
The low yields of certain groups of patients has also been found by other studies [9, 10]. There were six requests for CT in patients presenting with transient ischaemic attacks and all had normal scans. It may be that in a patient with a well documented transient ischaemic attack and full neurological recovery within 24 hours, anticoagulant therapy can be safely commenced without a scan [11, 12]. Similarly, scanning patients for progression of neurological signs following a stroke did not alter the further management of these patients. Sandercock et al.  found just one case of a |nonstroke' lesion in 77 patients with progressive neurological deficit. The high mortality in the group of patients in our study and the lack of new diagnostic information from CT scans raises doubt about the utility of scanning such patients.
The results of our study seem to suggest that computed tomography of the brain is a valuable first-line investigation in elderly patients presenting with signs atypical of stroke and unexplained confusion. Our findings reflect the non-specific nature of presenting clinical features of treatable neurological lesions in this age group and suggest that criteria applicable to younger patients for CT scanning should be modified for elderly subjects. Prospective studies are required to confirm this.
[1.] Sandercock P, Molyneux A, Warlow C. Value of computed tomography in patients with stroke: Oxfordshire Community Stroke Project. Br Med F 1985;290:193-7. [2.] Allen CMC, Harrison MJC, Wade DT. The management of acute stroke. Tunbridge Wells: Castle House Publications, 1988. [3.] Warlow CP. Cerebrovascular disease. In: Weatherall DJ, Ledingham JGG, Warrell DA, eds. Oxford textbook of medicine. 2nd edn. Oxford: Oxford University Press, 1987. [4.] Kaye JA, DeCarli, Luxenberg JS, Rapoport SI. The significance of age-related enlargement of the cerebral ventricles in healthy men and women measured by quantitative computed X-ray tomography. F Am Geriatr Soc 1992;40:225-31. [5.] Colgan J, Naguib, Levy R. Computed tomographic density numbers: a comparative study of patients with senile dementia and normal elderly controls. Neurology 1985;35:1316-20. [6.] Raskind R, Glover MB, Weiss SR. Chronic subdural hematoma in the elderly: a challenge in diagnosis and treatment. F Am Geriatr Soc 1972;20:330-4. [7.] Doherty DL. Posttraumatic cerebral atrophy as a risk factor for delayed acute subdural haemorrhage. Arch Phys Med Rehabil 1988;69:542-4. [8.] Roberts MA, Caird FI. The contribution of computerized tomography to the differential diagnosis of confusion in elderly patients. Age Ageing 1990;19:50-6. [9.] Larson EB, Omenn GS, Loop JW. Computer tomography in patients with cerebrovascular disease: impact of a new technology on patient care. Am F Roentgenol 1978;131:35-40. [10.] Hazelton AE, Earnest MP. Impact of computed tomography on stroke management and outcome. Arch Intern Med 1987;147:217-20. [11.] UK-TIA Study Group: United Kingdom transient ischaemic attack (UK-TIA): interim results. Br Medg F 1988;296:316-20. [12.] Sandercock P. Management of transient ischaemic attacks. Hosp Update 1990;16:725-33.
G. Brown, M. Warren, J. E. Williams, E. J. Adam Department of Diagnostic Radiology,
J.A. Coles Department of Geriatric Medicine,
St George's Hospital, London SW17 OQT
Received 7 August 1992
|Printer friendly Cite/link Email Feedback|
|Author:||Brown, Gina; Warren, Mary; Williams, J.E.; Adam, E.J.; Coles, J.A.|
|Publication:||Age and Ageing|
|Date:||Jul 1, 1993|
|Previous Article:||Ageing as a global issue.|
|Next Article:||A study of nutritional deficits of long-stay geriatric patients.|