Management of cognitive impairment of vascular origin

A postal questionnaire concerning the management of cognitive impairment of vascular origin was completed by 88 psychogeriatricians and 100 geriatricians. Aspirin was considered an important part of management, and clinically indicated with: a history, or clinical signs of stroke; computerised tomography (CT) scan changes suggestive of cerebrovascular disease; and atrial fibrillation. Advanced dementia was seen as a contraindication for treatment. Eighty per cent of responders indicated they would institute antihypertensive therapy in an elderly patient with a history of raised blood pressure and cognitive impairment. The median lower limit of systolic blood pressure requiring treatment was loOmmHg, with a treatment target blood pressure of ISOmmHg.

EpidemiolÃ³gica! surveys and clinicopathological studies have established that cerebrovascular disease is one of the three main causes of dementia (Tomlinson et al, 1970;Rocca et al, 1991;OU et al, 1995). A variety of pathological processes may be involved, such as multiple cortical or lacunar infarcts and white matter ischaemia (Amar & Wilcock, 1996). There are now established preventative strategies for eld erly patients at increased risk of developing stroke: the treatment of hypertension in older adults (Medical Research Council Working Party, 1992), and aspirin therapy (Antiplatelet Trialist's Collaboration, 1994) have both been found to be beneficial. There is still very little research specific to the treatment of vascular dementia, but some evidence does exist to suggest that cognitive function may be improved in those with vascular dementia with control of systolic hy pertension and aspirin therapy (Meyer et al, 1986(Meyer et al, . 1989).

The study
We conducted a postal survey of psychogeria tricians and physicians examining the manage ment of patients with vascular-related cognitive impairment. The term 'cognitive impairment of probable vascular origin' was chosen in pre ference to vascular dementia to include those suffering with milder and focal impairments not classifiable as global dementia. A short two-sided questionnaire was circulated to 133 consultant psychiatrists and 195 consultant physicians working with the elderly within four health regions. Returned forms were identified by a number specific for each clinician. Those not responding after six weeks received a second mailing. Results are expressed as percentages, and analysed using SPSS for Windows (1993). Comparisons between psychiatrists and geria tricians were performed using /2 for categorical data, and two-tailed t-tests for interval data.

Findings
A total of 189 questionnaires (58%) were re turned. The response rate from psychogeriatric ians (88/133; 66%) was much higher than from geriatricians (100/195: 51%). Due to a damaged return, the speciality of one clinician could not be identified. The majority of clinicians (90: 48%) felt that cerebrovascular disease was responsible for less than 30% of patients they saw with cognitive impairment, and 72 (38%) for 30-50% of patients. Geriatricians were more likely to estimate a higher prevalence of probable vascu lar-related cognitive impairment than psychi atrists (/2=10.2. d.f.=2, P=0.006). One-third estimated 50-80% of patients with vascularrelated cognitive impairment were receiving aspirin, and only 21 (11%) thought that less than 10% were receiving aspirin. There were no significant differences between psychiatrists and geriatricians (/2=5.8, d.f.=5. P=0.33). Possible indications for aspirin therapy in a patient with cognitive impairment are considered in Table 1. There were high rates of potential aspirin therapy in patients with a history or clinical signs of stroke, and brain computed tomography sugges tive of cerebrovascular disease. Atrial fibrillation was considered an indication for aspirin by the majority of clinicians, though geriatricians were more likely to prescribe than psychiatrists (X2=13, d.f.=2, P=0.002). A history suggestive of 'mixed dementia' was also commonly associated with aspirin therapy (119; 63%). Table 2 con siders particular clinical situations which might preclude the use of aspirin. Most viewed severe end-stage dementia, history of peptic ulcer, and cerebral haemorrhage as contraindications, and would exercise caution when there is a history of iron deficiency anaemia, allergy and asthma. The most widely prescribed dose of aspirin was 75 mg (104: 55%). followed by 150mg (66; 35%).
One hundred and fifty-two responders (81%) felt they would institute anti-hypertensive ther apy in a patient with a history of raised blood pressure and cognitive impairment of probable vascular origin, 13 (7%) would not, and 23 (12%) were unsure. The median lower limit of systolic blood pressure considered pathological and meriting treatment in elderly people was 160 mmHg. The median target treatment systolic blood pressure was 150 mmHg. There was no significant difference between psychiatrists or physicians on their perception of a pathologically elevated blood pressure, or treatment target blood pressure. The majority of respondents (119/181: 65%) felt that age would influence their management of hypertension, many practi tioners deciding not to treat hypertension beyond the age of 80 unless severe. Diuretics (125/180: 69%), proved by far the most popular choice of drug for first line treatment of hypertension.

Comment
The survey showed a clear consensus that aspirin is generally seen as an important management strategy: history of stroke, transi ent ischaemic attack, atrial fibrillation, clinical signs of stroke, and CT evidence of infarct or perivascular shadowing are clearly viewed as indicators for aspirin. Aspirin is less effective than warfarin in reducing the risk of stroke in patients with atrial fibrillation, but appears a good compromise in cognitively impaired pa tients with fewer complications associated with treatment (Lip& Lowe, 1996). The dose of aspirin used by the Antiplatelet Trialists' Collaboration (1994) varied from 75-325 mg. and doses throughout this range seemed similarly effective in reducing vascular events in high-risk patients. However, the odds ratio for risk of hospitalisation from peptic ulcer bleed is lower for 75 mg (2.3), compared to 150mg (3.2), and 300 mg (3.9) (Weil et al, 1995). A history of peptic ulcer and cerebral haemorrhage were seen by the majority as a contraindication for aspirin therapy, and the presence of an iron deficiency anaemia and asthma as reasons for caution. We asked clinicians to consider if they would prescribe aspirin to a patient with a severe 'end-stage' dementia, such as a patient who was immobile. There was a clear consensus that in these clinical circumstances it would be inappropriate to prescribe.
The evidence for treating hypertension in the elderly is now strong, with studies consistently showing reduction in the risk of stroke (Sander son, 1996). Any benefit of treating those over 80 years of age has yet to be determined; though in a study involving a proportion of patients over this age, Starr et ai (1996) found a higher mortality in healthy old people with a casual systolic blood pressure above ISOmmHg. Meyer et al (1986) found improved cognition in patients with vascular dementia when systolic blood pressure was controlled within the range 135-150 mmHg.
Anticoagulation with warfarin is not without problems in the elderly, and benefits and risk need to be carefully considered. Not surprisingly geriatricians were more likely to consider pre scribing warfarin. Adequate supervision of the patient, availability of a monitoring service, risk of falls, and degree of impairment are all important factors.
A questionnaire of this nature is always open to bias, namely responders may take more of an interest in the area of enquiry compared to nonresponders. In addition, clinicians' actual prac tice may in reality differ from their response to questionnaires (Jones et al, 1990). We found that the majority of clinicians were in favour of pharmacological preventative strategies for peo ple with cognitive impairment of vascular origin.