630
25. Greenwald BD, Harpaz N, Yin J, et al. Loss of heterozygosity affecting the p53, Rb, and mcc/apc tumor suppressor gene loci in dysplastic and cancerous ulcerative-colitis. Cancer Res 1992; 52: 741-45. 26. D’Amico D, Carbone DP, Johnson BE, Meltzer SJ, Minna JD. Polymorphic sites within the MCC and APC loci reveal very frequent loss of heterozygosity in human small-cell lung-cancer. Cancer Res 1992; 52: 1996-999. 27. Bodmer WF. Hereditary colorectal cancer: a commentary. In: Utsunomiya J, Lynch HT, eds. Hereditary colorectal cancer. Tokyo: Springer-Verlag, 1990: 37-42.
28. Bodmer WF, McMichael A. Introduction. In: Bodmer WF, McMichael A, eds. A new look at tumour immunology. Cancer surveys vol 13. New York: Cold Spring Harbor Laboratory press, 1992: 1-4. 29. Bishop DT, Cannon-Albright LA, Burt RW, Skolnick MH. Pedigree
30.
analysis of susceptibility to colonic adenomas. In: Utsunomiya J, Lynch HT, eds. Hereditary colorectal cancer. Tokyo: SpringerVerlag, 1990: 423-30. Leibovitz A, Stinson JC, McCombs III WB, McCoy CE, Mazur KC, Mabry ND. Classification of human colorectal adenocarcinoma cell lines. Cancer Res 1976; 36: 4562-69.
Risk of cardiac events in atypical transient ischaemic attack or minor stroke
Proposed guidelines for the diagnosis of transient ischaemic attack (TIA) involve interpretation of symptoms, so it can be very difficult to distinguish a TIA from other disorders, such as migraine, epilepsy, syncope, or neurosis. Atypical cerebral and visual events may be classified as TIA. To see whether TIA or stroke patients with atypical cerebral or visual symptoms are at high or low risk of cardiac complications, we prospectively followed 572 patients (entered into the Dutch multicentre TIA trial) with a diagnosis of TIA or minor ischaemic stroke, but whose symptoms did not fully accord with internationally accepted criteria. We compared their outcome with that of 2555 other TIA or stroke patients in the trial, who had unequivocal symptoms; all patients were treated with aspirin. During mean follow-up of 2·6 years the risk of a major vascular event did not differ between the groups (14·5% in patients with atypical symptoms vs 15·1% of patients with typical attacks). Patients with atypical attacks had a lower risk of stroke (5·6% vs 9·4%, hazard ratio 0·6, 95% confidence interval 0·4-0·9) and a higher risk of a major cardiac event (8·4% vs 5·9%, 1·4, 1·0-2·0) than did patients with typical attacks. These differences could not be explained by differences in cardiac risk factors, and were independent of minor discrepancies in baseline characteristics between the groups. A heavy or tired feeling in one or two limbs was the only atypical symptom associated with cerebral rather than cardiac events (ratio cardiac/cerebral events 0·8). For all other atypical symptoms cardiac events were about twice as common as cerebral events (range
1·3-2·5). Our findings suggest that TIA or minor stroke patients with atypical symptoms may have symptomatic heart disease, especially cardiac
diagnosis of TIA entail interpretation of symptoms, distinguishing a TIA from other disorders, such as migraine, epilepsy, syncope, or even neurosis,2-4 remains difficult. Unusual attacks of visual blurring have been associated with extensive atherosclerotic disease of the carotid arteries,5,6 but atherosclerosis is common also in symptom-free elderly subjects.’ On the other hand, brief and transient attacks of complete blindness were predictors of disabling stroke.8 In patients with non-focal attacks of isolated dizziness, giddiness, or "wooziness", postural hypotension or cardiac arrhythmias may be misdiagnosed as TIAs. Although repeated jerking during the attack suggests an epileptic cause, involuntary movements, described as "limb shaking", have been seen in patients with extensive atherosclerotic lesions in the carotid arteries.9 Thus, despite the recommended criteria for the diagnosis of TIA, the neurologist still faces a diagnostic "grey area".2,3,10 The prognosis of patients with atypical attacks is uncertain. Heyden et alll showed, in a general population, a lower overall mortality, and also a lower stroke rate, in patients with "uncertain TIAs" than in those with definite TIAs, but the number of patients was small and a relation between symptoms and further cardiac or cerebral events was not especially sought. In a retrospective study, in which control patients with a definite TIA or minor stroke were not concurrent, we12 suggested that atypical TIAs were later associated with cardiac rather than cerebral events. To assess whether TIA or stroke patients with atypical cerebral or visual symptoms have an exceptionally high risk of cardiac complications, we have compared outcome for patients with a diagnosis of TIA or minor ischaemic stroke whose symptoms did not fully accord with the internationally accepted criteria,1 with that for TIA or minor stroke patients who had unequivocal symptoms. 13,14
Patients and methods Patients All patients were entered into the Dutch TIA trial, a multicentre clincial trial, whose background, design, and results have been
arrhythmia. Introduction
Diagnosis of transient ischaemic attack (TIA) rests entirely on the neurologist’s skill in questioning the patient and in deciding whether the symptoms suggest a risk of future stroke. Since the proposed guidelines for the
ADDRESSES. Department of Neurology, University Hospital Rotterdam (P J Koudstaal, MD, J. C. van Latum, MD), University Department of Neurology, Utrecht (A. Algra, MD, L. J. Kappelle, MD, Prof J van Gijn, FRCPE), and Department of Cardiology, Holy Hospital Vlaardingen, the Netherlands (G. A. M. Pop, MD). Correspondence to Dr Peter J. Koudstaal, Department of Neurology, University Hospital Rotterdam Dijkzigt, 40 Dr Molewaterplein, 3015 GD Rotterdam, the Netherlands
631
TABLE I-OUTCOME IN PATIENTS WITH TYPICAL OR ATYPICAL SYMPTOMS
i
HR= hazard rat!0,C!= confidence interval,
I
I
I
MI = myocardial infarction
described elsewhere.13,14 Briefly, patients with one or more cerebral ischaemic attacks were randomised to receive 30 mg or 283 mg aspirin, and, by means of a factorial design, about half of them were also randomised to receive 50 mg atenolol or placebo. Patients with minor stroke were eligible provided they were independent in most activities of daily living (grade 3 or better on the modified Rankin scale)." Patients with a presumed source of embolism in the heart were excluded, as were patients with vasculitis or blood
dyscrasias. Between 1986 and 1989 a total of 3150 patients were randomised. 23 patients whose diagnosis was evidently wrong at the time of randomisation (eg, cerebral tumour or multiple sclerosis) were excluded from the present analysis. Of the remaining patients, 997 had had a TIA (symptoms completely reversible within 24 hours) as their presenting attack and 2130 had had a minor stroke (symptoms for longer than 24 hours). In both categories, patients could have had a non-disabling stroke previously. The patients were followed up by their neurologist at 4-monthly intervals. No patient was lost to
follow-up.
Neurological symptoms For recording the history a check-list, in which the possible symptoms were described in simple language, was used.3 The list contained detailed multiple-choice questions about the nature of the symptoms and about the time course, including mode of onset, synchronicity of symptoms, duration of attack, mode of disappearance, and number of attacks. For each patient we assessed whether the neurological symptoms were fully compatible with the internationally accepted criteria for a TIA,’ using a specially designed computer program. The performance of this computer analysis was manually checked in a random sample of 300 patients. TIAs and minor strokes were judged atypical if the patient had had one or more of the following symptoms: disturbances of vision in one or both eyes consisting of flashes, lines, objects, distorted view, tunnel vision, image moving on change of posture; alteration of muscle strength consisting of tired or heavy sensation in one or more limbs, either unilateral or bilateral; sensory symptoms alone (unilateral or bilateral) or a gradual spread of sensory symptoms; brainstem symptoms and coordination difficulties consisting of isolated disorder of swallowing or articulation, double vision, dizziness, or uncoordinated movements; accompanying symptoms including unconsciousness, limb jerking, tingling of limbs or lips, disorientation, amnesia.
Baseline characteristics In addition to the neurological history, vascular risk factors and vascular diseases were recorded. 13,11 Computed tomography (CT) of the brain was required in all patients apart from those with transient monocular blindness. We also collected data from a standard 12-lead electrocardiogram (ECG). All CT scans and ECGs were reviewed by two neurologists and two cardiologists, respectively, without knowledge of patient data.
Incidence of major vascular events (A), fatal or non-fatal stroke (B), and cardiac events (C) in patients with typical or
atypical neurological symptoms.
Vertical lines = 95% confidence limits
Outcome events The definition of vascular death included sudden death
(unexpected cardiac death within an hour of onset of symptoms, or within 24 hours if the patient was unexpectedly found dead), or death from stroke, myocardial infarction, congestive heart failure, or systemic bleeding. For diagnosis of non-fatal stroke, appropriate clinical features persisting for more than 24 hours had to correspond to a fresh infarct or haemorrhage on a repeat CT scan. If CT showed no changes or was not available, an increase in disability of at least one grade on the modified Rankin scale was required for the diagnosis of stroke. Myocardial infarction had to be documented by at least two of the following characteristics: history of chest discomfort, changes in specific enzymes more than twice the upper range of normal, or development of Q-waves on standard 12-lead ECG. All outcome events were independently classified by at least three
assessors.
Data
analysis
The occurrence of outcome events was compared in terms of the hazard ratio (HR) which may be interpreted as a relative risk-ie,
632
TABLE II-BASELINE CHARACTERISTICS OF PATIENTS WITH ATYPICAL OR TYPICAL SYMPTOMS
In a separate analysis we assessed which of the atypical symptoms in particular implied an increased risk of cardiac events. In all patients with atypical symptoms the cardiac to cerebral events ratio was 1-5 compared with 0-6 in patients with typical attacks. A heavy or tired feeling in one or two limbs was the only atypical symptom associated with stroke rather than cardiac events (ratio 0-8). For all other atypical symptoms cardiac events were about twice as common as cerebral events (ratio ranging from 1-3 to 2-5).
Discussion We have shown that patients with symptoms that resemble TIA or minor stroke but that do not accord with internationally accepted criteria have a low risk of stroke but a high risk of major cardiac events compared with patients with "typical" TIAs or minor strokes. We realise that our results apply to a selected group of patients with suspected cerebral ischaemia. These results could not be explained by differences in cardiovascular risk factors and were independent of minor differences in other baseline characteristics between the groups. Our findings confirm the results of Heyden et al,l1 as well as those of our earlier retrospective study, in which the control group was not concurrent.12 In a more extensive study of risk factors for major vascular events in patients entered into the Dutch TIA trial, atypical neurological symptoms proved independent from other prognostic factors BP=blood pressure,
*p < 0 05,
PVC=premature ventricular complex, Ct=carchothorac!C.
†p
25. Greenwald BD, Harpaz N, Yin J, et al. Loss of heterozygosity affecting the p53, Rb, and mcc/apc tumor suppressor gene loci in dysplastic and cancerous ulcerative-colitis. Cancer Res 1992; 52: 741-45. 26. D’Amico D, Carbone DP, Johnson BE, Meltzer SJ, Minna JD. Polymorphic sites within the MCC and APC loci reveal very frequent loss of heterozygosity in human small-cell lung-cancer. Cancer Res 1992; 52: 1996-999. 27. Bodmer WF. Hereditary colorectal cancer: a commentary. In: Utsunomiya J, Lynch HT, eds. Hereditary colorectal cancer. Tokyo: Springer-Verlag, 1990: 37-42.
28. Bodmer WF, McMichael A. Introduction. In: Bodmer WF, McMichael A, eds. A new look at tumour immunology. Cancer surveys vol 13. New York: Cold Spring Harbor Laboratory press, 1992: 1-4. 29. Bishop DT, Cannon-Albright LA, Burt RW, Skolnick MH. Pedigree
30.
analysis of susceptibility to colonic adenomas. In: Utsunomiya J, Lynch HT, eds. Hereditary colorectal cancer. Tokyo: SpringerVerlag, 1990: 423-30. Leibovitz A, Stinson JC, McCombs III WB, McCoy CE, Mazur KC, Mabry ND. Classification of human colorectal adenocarcinoma cell lines. Cancer Res 1976; 36: 4562-69.
Risk of cardiac events in atypical transient ischaemic attack or minor stroke
Proposed guidelines for the diagnosis of transient ischaemic attack (TIA) involve interpretation of symptoms, so it can be very difficult to distinguish a TIA from other disorders, such as migraine, epilepsy, syncope, or neurosis. Atypical cerebral and visual events may be classified as TIA. To see whether TIA or stroke patients with atypical cerebral or visual symptoms are at high or low risk of cardiac complications, we prospectively followed 572 patients (entered into the Dutch multicentre TIA trial) with a diagnosis of TIA or minor ischaemic stroke, but whose symptoms did not fully accord with internationally accepted criteria. We compared their outcome with that of 2555 other TIA or stroke patients in the trial, who had unequivocal symptoms; all patients were treated with aspirin. During mean follow-up of 2·6 years the risk of a major vascular event did not differ between the groups (14·5% in patients with atypical symptoms vs 15·1% of patients with typical attacks). Patients with atypical attacks had a lower risk of stroke (5·6% vs 9·4%, hazard ratio 0·6, 95% confidence interval 0·4-0·9) and a higher risk of a major cardiac event (8·4% vs 5·9%, 1·4, 1·0-2·0) than did patients with typical attacks. These differences could not be explained by differences in cardiac risk factors, and were independent of minor discrepancies in baseline characteristics between the groups. A heavy or tired feeling in one or two limbs was the only atypical symptom associated with cerebral rather than cardiac events (ratio cardiac/cerebral events 0·8). For all other atypical symptoms cardiac events were about twice as common as cerebral events (range
1·3-2·5). Our findings suggest that TIA or minor stroke patients with atypical symptoms may have symptomatic heart disease, especially cardiac
diagnosis of TIA entail interpretation of symptoms, distinguishing a TIA from other disorders, such as migraine, epilepsy, syncope, or even neurosis,2-4 remains difficult. Unusual attacks of visual blurring have been associated with extensive atherosclerotic disease of the carotid arteries,5,6 but atherosclerosis is common also in symptom-free elderly subjects.’ On the other hand, brief and transient attacks of complete blindness were predictors of disabling stroke.8 In patients with non-focal attacks of isolated dizziness, giddiness, or "wooziness", postural hypotension or cardiac arrhythmias may be misdiagnosed as TIAs. Although repeated jerking during the attack suggests an epileptic cause, involuntary movements, described as "limb shaking", have been seen in patients with extensive atherosclerotic lesions in the carotid arteries.9 Thus, despite the recommended criteria for the diagnosis of TIA, the neurologist still faces a diagnostic "grey area".2,3,10 The prognosis of patients with atypical attacks is uncertain. Heyden et alll showed, in a general population, a lower overall mortality, and also a lower stroke rate, in patients with "uncertain TIAs" than in those with definite TIAs, but the number of patients was small and a relation between symptoms and further cardiac or cerebral events was not especially sought. In a retrospective study, in which control patients with a definite TIA or minor stroke were not concurrent, we12 suggested that atypical TIAs were later associated with cardiac rather than cerebral events. To assess whether TIA or stroke patients with atypical cerebral or visual symptoms have an exceptionally high risk of cardiac complications, we have compared outcome for patients with a diagnosis of TIA or minor ischaemic stroke whose symptoms did not fully accord with the internationally accepted criteria,1 with that for TIA or minor stroke patients who had unequivocal symptoms. 13,14
Patients and methods Patients All patients were entered into the Dutch TIA trial, a multicentre clincial trial, whose background, design, and results have been
arrhythmia. Introduction
Diagnosis of transient ischaemic attack (TIA) rests entirely on the neurologist’s skill in questioning the patient and in deciding whether the symptoms suggest a risk of future stroke. Since the proposed guidelines for the
ADDRESSES. Department of Neurology, University Hospital Rotterdam (P J Koudstaal, MD, J. C. van Latum, MD), University Department of Neurology, Utrecht (A. Algra, MD, L. J. Kappelle, MD, Prof J van Gijn, FRCPE), and Department of Cardiology, Holy Hospital Vlaardingen, the Netherlands (G. A. M. Pop, MD). Correspondence to Dr Peter J. Koudstaal, Department of Neurology, University Hospital Rotterdam Dijkzigt, 40 Dr Molewaterplein, 3015 GD Rotterdam, the Netherlands
631
TABLE I-OUTCOME IN PATIENTS WITH TYPICAL OR ATYPICAL SYMPTOMS
i
HR= hazard rat!0,C!= confidence interval,
I
I
I
MI = myocardial infarction
described elsewhere.13,14 Briefly, patients with one or more cerebral ischaemic attacks were randomised to receive 30 mg or 283 mg aspirin, and, by means of a factorial design, about half of them were also randomised to receive 50 mg atenolol or placebo. Patients with minor stroke were eligible provided they were independent in most activities of daily living (grade 3 or better on the modified Rankin scale)." Patients with a presumed source of embolism in the heart were excluded, as were patients with vasculitis or blood
dyscrasias. Between 1986 and 1989 a total of 3150 patients were randomised. 23 patients whose diagnosis was evidently wrong at the time of randomisation (eg, cerebral tumour or multiple sclerosis) were excluded from the present analysis. Of the remaining patients, 997 had had a TIA (symptoms completely reversible within 24 hours) as their presenting attack and 2130 had had a minor stroke (symptoms for longer than 24 hours). In both categories, patients could have had a non-disabling stroke previously. The patients were followed up by their neurologist at 4-monthly intervals. No patient was lost to
follow-up.
Neurological symptoms For recording the history a check-list, in which the possible symptoms were described in simple language, was used.3 The list contained detailed multiple-choice questions about the nature of the symptoms and about the time course, including mode of onset, synchronicity of symptoms, duration of attack, mode of disappearance, and number of attacks. For each patient we assessed whether the neurological symptoms were fully compatible with the internationally accepted criteria for a TIA,’ using a specially designed computer program. The performance of this computer analysis was manually checked in a random sample of 300 patients. TIAs and minor strokes were judged atypical if the patient had had one or more of the following symptoms: disturbances of vision in one or both eyes consisting of flashes, lines, objects, distorted view, tunnel vision, image moving on change of posture; alteration of muscle strength consisting of tired or heavy sensation in one or more limbs, either unilateral or bilateral; sensory symptoms alone (unilateral or bilateral) or a gradual spread of sensory symptoms; brainstem symptoms and coordination difficulties consisting of isolated disorder of swallowing or articulation, double vision, dizziness, or uncoordinated movements; accompanying symptoms including unconsciousness, limb jerking, tingling of limbs or lips, disorientation, amnesia.
Baseline characteristics In addition to the neurological history, vascular risk factors and vascular diseases were recorded. 13,11 Computed tomography (CT) of the brain was required in all patients apart from those with transient monocular blindness. We also collected data from a standard 12-lead electrocardiogram (ECG). All CT scans and ECGs were reviewed by two neurologists and two cardiologists, respectively, without knowledge of patient data.
Incidence of major vascular events (A), fatal or non-fatal stroke (B), and cardiac events (C) in patients with typical or
atypical neurological symptoms.
Vertical lines = 95% confidence limits
Outcome events The definition of vascular death included sudden death
(unexpected cardiac death within an hour of onset of symptoms, or within 24 hours if the patient was unexpectedly found dead), or death from stroke, myocardial infarction, congestive heart failure, or systemic bleeding. For diagnosis of non-fatal stroke, appropriate clinical features persisting for more than 24 hours had to correspond to a fresh infarct or haemorrhage on a repeat CT scan. If CT showed no changes or was not available, an increase in disability of at least one grade on the modified Rankin scale was required for the diagnosis of stroke. Myocardial infarction had to be documented by at least two of the following characteristics: history of chest discomfort, changes in specific enzymes more than twice the upper range of normal, or development of Q-waves on standard 12-lead ECG. All outcome events were independently classified by at least three
assessors.
Data
analysis
The occurrence of outcome events was compared in terms of the hazard ratio (HR) which may be interpreted as a relative risk-ie,
632
TABLE II-BASELINE CHARACTERISTICS OF PATIENTS WITH ATYPICAL OR TYPICAL SYMPTOMS
In a separate analysis we assessed which of the atypical symptoms in particular implied an increased risk of cardiac events. In all patients with atypical symptoms the cardiac to cerebral events ratio was 1-5 compared with 0-6 in patients with typical attacks. A heavy or tired feeling in one or two limbs was the only atypical symptom associated with stroke rather than cardiac events (ratio 0-8). For all other atypical symptoms cardiac events were about twice as common as cerebral events (ratio ranging from 1-3 to 2-5).
Discussion We have shown that patients with symptoms that resemble TIA or minor stroke but that do not accord with internationally accepted criteria have a low risk of stroke but a high risk of major cardiac events compared with patients with "typical" TIAs or minor strokes. We realise that our results apply to a selected group of patients with suspected cerebral ischaemia. These results could not be explained by differences in cardiovascular risk factors and were independent of minor differences in other baseline characteristics between the groups. Our findings confirm the results of Heyden et al,l1 as well as those of our earlier retrospective study, in which the control group was not concurrent.12 In a more extensive study of risk factors for major vascular events in patients entered into the Dutch TIA trial, atypical neurological symptoms proved independent from other prognostic factors BP=blood pressure,
*p < 0 05,
PVC=premature ventricular complex, Ct=carchothorac!C.
†p
