Patients With Transient Ischemic Attack Do Not Need To Be Admitted to Hospital for Urgent Evaluation and Treatment
The essence of the assessment of a patient with suspected transient ischemic attack (TIA) is to confirm the diagnosis, identify and treat the cause, and start effective secondary prevention to prevent a disabling or fatal vascular event. In general, admission to hospital is required for patients who are medically unstable, those who have become dependent or require complex care that is unfeasible or unavailable elsewhere. You do not need to be admitted to hospital because of a TIA, but this is often done because of the absence of an alternative. Many TIA clinics now offer a “one-stop” service for which the patient is assessed, investigated (or investigated before the appointment), and given results at the same session.1 The chief problem of such a clinic is that patients may experience a completed stroke (or other vascular disaster) before being seen, so delays must be eliminated.2,3 Solutions include immediate access to regular clinics (by accepting telephone, email, or faxed referrals) and better public recognition of the seriousness of a TIA or minor stroke.
Confirming the diagnosis or identifying one of the many TIA mimics is an important aspect of a TIA clinic because about one third of referrals do not have cerebrovascular disease.4 In addition, the wide variety of neurological and general medical conditions that mimic TIA provide an invaluable resource for medical training. Appropriately trained personnel are vital, and the more senior personnel (including consultant specialists) who run TIA clinics are preferred over the typically less experienced staff initially assessing patients in the emergency room.
Once a definite diagnosis of TIA is made, examination and investigation is required to identify the cause. This should include blood tests and ECG for all patients, brain imaging for most, and selected use of echocardiography (transthoracic and transesophageal) and carotid scanning, and all of these can be done as an outpatient. Weight, blood pressure, glucose, cholesterol, and ECG testing can be done immediately at the bedside. A coordinated approach can facilitate access to magnetic resonance scanning (including gradient echo sequences to identify small bleeds in those patients who inevitably present late with minor stroke symptoms5), computed tomography scanning to exclude a tumor (for those with multiple attacks), and echocardiography to help identify a cardiac or intrathoracic large vessel cause.
A patient with a TIA (or minor stroke) needs specialist advice and management, and this is often best done in the privacy of a clinic environment, especially if relatives are routinely available. Patients with cerebrovascular disease are increasingly elderly6 and commonly have significant comorbidity that will influence management plans.7 Patients need to be told their diagnosis and advised on lifestyle management and medication. Patients can be anticoagulated as an outpatient (many such services are now available), and aspirin or other antithrombotic should be started for those in sinus rhythm (do not forget the immediate loading dose required for aspirin and clopidogrel). Blood pressure and cholesterol lowering can be started (stop smoking, exercise regularly, reduce dietary salt and fat, and commence a diuretic, angiotensin-converting enzyme inhibitor and statin over the following few weeks) as dictated by tests and individual circumstances. Patients with a symptomatic tight carotid stenosis need to be fast-tracked to the local vascular team.
Keeping patients out of hospitals cannot only be cost-effective but often provides the best service to patients, satisfies our political masters, and helps patients avoid those iatrogenic disasters of modern large hospitals (deep venous thrombosis, epidemic norovirus, methicillin resistant staphylococcus aureus, and deconditioning of frail older people). An outpatient service that offers a prompt specialized assessment is also a superb environment for training, audit, and research.
Finally, it is worth returning to the major criticism of a TIA clinic, namely whether serious or fatal strokes would occur in the waiting time between referral and being seen. Delays must be eliminated. However, we also need to consider why these events occur and whether hospital (as opposed to community) treatment would really make a difference? Our proven interventions include antithrombotic medication, which can be started by the referring doctor. Only a minority of TIA patients have tight symptomatic carotid stenosis requiring an early operation, but would a delay of a few days be significant? Statins probably need many months of treatment to have an effect8 (acute blood pressure and cholesterol-lowering trials are the focus of new trials). Anticoagulation for patients with atrial fibrillation often needs considerable thought, and antiplatelet therapy for a few days may be reasonable until a considered judgment is made about longer-term treatment.
Patients will continue to present late after TIA, some patients will refuse hospital admission, and many may prefer an urgent outpatient assessment over a hospital admission, and thus, patients do not have to be admitted to hospital for their TIA.
Professor Lindley is supported by an infrastructure grant from NSW Health.
I would like to thank Dr Peter Landau for his helpful comments during the preparation of this article.
Coull AJ, Lovett JK, Rothwell PM. Population based study of early risk of stroke after transient ischaemic attack or minor stroke: implications for public education and organization of services. BMJ. 2004; 328: 326.
Widjaja E, Salam SN, Griffiths PD, Kamara C, Doyle C, Venables GS. Is the rapid assessment stroke clinic rapid enough in assessing transient ischaemic attack and minor stroke? J Neurol Neurosurg Psychiatry. 2005; 76: 146–147.
Martin PJ, Young G, Enevoldson TP, Humphrey PR. Overdiagnosis of TIA and minor stroke: experience at a regional neurovascular clinic. Q J Med. 1997; 90: 759–763.
Wardlaw JM, Keir SL, Dennis MS. The impact of delays in computed tomography of the brain on the accuracy of diagnosis and subsequent management in patient with minor stroke. J Neurol Neurosurg Psychiatry. 2003; 74: 77–81.