MB McFadyen
Neurology Outpatient Department, Institute of Neurological Sciences, Southern General Hospital, Glasgow.
Correspondence to: Dr M.B.McFadyen, Department of Clinical Neurophysiology, Regional Neuroscience Centre, Newcastle General Hospital Westgate Road, Newcastle
E-mail: morag.mcfadyen@btinternet.com
SMJ 2004 49(4): 126-130
Background: Despite high life time prevalence of first seizures there is a dearth of information relating to uptake and efficient use of resources for first seizure services. Aims: To determine referral pattern, epilepsy and non-epilepsy diagnoses, investigations performed, follow-up appointments generated and resource utilisation at a first seizure clinic. Methods: A prospective study of 200 consecutive attendees was performed over a 16 month period at a first seizure clinic in the Institute of Neurological Sciences, Glasgow. Results: 139(69.5%) patients were general practitioner and 61(30.5%) hospital referrals. Fifty-six (28%) patients were newly diagnosed with epilepsy and 26(13%) with an unprovoked first seizure. Alcohol and/or drug abuse were contributing factors in 23 patients or 19.5% of all seizure cases. Twenty-one (35%) of 62 EEGs contained epileptiform activity which assisted with an epilepsy diagnosis. Overall 22(29%) of 77 CT head scans disclosed abnormalities and 20(36%) of 56 in those with an index event of seizure. Brain tumours were identified in four patients or 5% of all patients who underwent CT scanning. All patients with an epilepsy diagnosis were offered follow up with a consultant neurologist and epilepsy nurse specialist. Late cancellation and non-attendence rates were 35-40% for initial clinic appointments and 16% for first seizure clinic review appointments. Conclusion: With provision of fast track, easy access first seizure services, individuals may be more likely to both attend and present earlier with fewer seizure events at first contact.
Key words: First seizure clinic, diagnoses, resource utilisation
Non febrile, single, unprovoked seizures have a lifetime prevalence between two and five% and their incidence in European studies ranges from 26 to 70 individuals per 100,000. The epilepsies have been recognised for over 4,000 years and with more than 50 million people worldwide, 6 million Europeans and 400,000 people in the United Kingdom affected, they are the most common group of serious neurological conditions. In developed societies their incidence ranges from 40 to 70 per 100,000 persons per year.1-3 Their prevalence is about five to 10 cases per 1,000 individuals, excluding febrile convulsions, single seizures and inactive cases.
There is no satisfactory test for the diagnosis of a single seizure, epilepsy and other paroxysmal disorders that can be applied when the patient is between attacks. In the great majority of cases, therefore, the diagnosis depends on clinical skills. Recent studies have suggested that there is a substantial error rate.4-6 Misidentification of the nature of attacks, compounded by failures in obtaining eyewitness accounts and misinterpretation of EEG reports by clinicians are highlighted as diagnostic errors made in up to a quarter of all patients seen. These problems have given rise to the recommendations, in the revised Scottish Intercollegiate Guideline Network (SIGN), that patients presenting with possible epileptic seizures should be assessed early at first seizure clinics, by doctors specialising in epilepsy.7
Whilst epidemiological studies may provide an estimate of the number of patients per year who may present with epilepsy from a given base population, it is not clear what resource provision needs to be made for first seizure services. In diagnosing epilepsy, inevitably a proportion of patients with other paroxysmal disorders will be seen and diagnosed. To evaluate the patient load for a given population, therefore, the summation of epileptic and non-epileptic seizure patients attending will be required. If clinic attendance rates can be established it should then be relatively easy to determine medical staffing requirements for outpatient first seizure services. However it is still uncertain what may be the implications of increased demand for investigations, such as neuroimaging, that such clinics may generate. In addition more information is required on the frequency of return clinic appointments and the likely uptake of services.
For these reasons a prospective audit of the case mix, epilepsy and non epilepsy diagnoses and resource utilisation of a first seizure clinic was undertaken. Referral patterns and event numbers were also recorded at first attendance.
The first seizure clinic was at the West of Scotland Regional Epilepsy Service, based in the Southern General Hospital in Glasgow. The audit was approved by the trust. Patients were 200 consecutive attendees seen by the same doctor and then discussed with or seen by a consultant neurologist specialising in epilepsy. Data were collected continuously over a 16 month period from March 2001 to June 2002 from patients referred with a diagnosis of possible first seizure or epilepsy. Clinical and demographic information was collected and recorded at the time of the initial consultation. Further data were collected during the course of the diagnostic process.
Patient characteristics and referral pattern
Of the 200 consecutive attendees from eight nationalities, the majority (71%) came from Greater Glasgow. The remaining 29% came from seven other Scottish health boards. One-hundred and thrity-nine(69.5%) patients were general practitioner referrals with tertiary referrals (61 patients, 30.5%) coming predominantly from general physicians (19.5%) and cardiologists (3.5%). Of the remainder, 6(3%) patients were referrals from Accident and Emergency, three (1.5%) each from Neurology/Neurourgery and Psychiatry, two(1%) from Ear Nose and Throat and one(0.5%) from Occupational Health. There were 116(58%) male and 84(42%) female patients (one of whom was pregnant) ranging in ages from 13 to 80. The cohort median age was 31 years (interquartile range 22-47). A sub-group of 13 patients (6.5%) had learning difficulties. Their median age was 27 years (interquartile range 19-51).
Forty-eight (24%) patients had experienced a single event at the time of first attendance. Twenty-six(13%) had a first unprovoked seizure. With 22 males and four females this group had a statistically significant (chi-square, p<0.005) preponderance of male attendees. The group median age was 34 years (interquartile range 22-44). Sixty-eight (34%) patients experiencing recurrent unprovoked seizures were identified and diagnosed with epilepsy. The group median age was 27.5 years (interquartile range 20-42) and with a male:female ratio of 36:32 there was no detectable sex bias. Patients who, in addition to a first tonic clonic seizure, described other seizure types, such as absences and matinal myoclonus, were diagnosed with epilepsy.
From a total of 94 patients with unprovoked seizures, 26 (27.6%) experienced one, 13(13.8%) two and eight (8.5%) three seizures. Thirty (31.9%) individuals had experienced multiple seizures (>20) before first contact.
Information gathering
In response to written requests, 121(60.5%) patients brought first hand eyewitnesses to their first, nine (4.5%) to their second and two (1%) to their third clinic appointments. Telephone contact elicited a further 29 eyewitness accounts whilst three patients without eyewitneses had typical events at the clinic. For 36(18%) individuals eyewitness accounts were unobtainable. First hand eyewitness accounts were obtained in 24 of 26 (92%) and 62 of 68(91%) of those with a first unprovoked seizure or epilepsy respectively, compared with two of 12 (17%) with an uncertain diagnosis.
Diagnoses
A total of 118 patients (59%) received a diagnosis of seizure. 26 individuals had an unprovoked first seizure defined as definite or highly probable in 22 cases and possible in four. 28% of all attendees or 56 of the 68 individuals with epilepsy were either newly diagnosed or had a recent diagnosis confirmed, with 52 definite or highly probable and four possible cases. A further seven individuals had a recurrence of seizure(s) following a prolonged seizure free period or inactive epilepsy. Five patients had established active epilepsy and were referred for reasons such as assessment of medically refractory epilepsy, consideration for the pre-surgical evaluation programme or transfer of care. These cases had either mesial temporal sclerosis, learning difficulties and / or cerebral palsy and all had suffered multiple seizures. Diagnostic data for all patients are presented in Table I.
Computerised T Tomography omography (CT)
Although SIGN guidelines clearly stipulate that Magnetic Resonance Imaging (MRI) is the modality of choice for brain imaging patients with epilepsy (MRI will detect lesions such as small tumours, vascular malformations and cortical dysplasias that may be missed by CT) 7, at the time of this audit MRI scans were not routinely available within the local National Health System for our first seizure clinic patients .Thus to circumvent the longer MRI waiting time (up to eight months during the audit period) CT head scans (waiting time 0-7 weeks) were requested initially. Seventy-seven (38.5%) patients from the clinic underwent computerized tomography (CT) brain scanning. CT head scans identified abnormalities in 22 (29%) cases and these are defined in Table II. Fifty-six scans (73%) were performed on patients with an index event of seizure and 20 (36%) found to be abnormal. In the group with a normal CT scan (n = 55), 36 had an index event of seizure(s) and 19 had other diagnoses. Referral of patients for CT scans at the time of this audit revealed a significantly greater number of abnormalities (chi-square, p<0.05) in the sub-group of patients with an index event of seizure (20 of 56) than those with other non-seizure related presentations (two of 21). CT disclosed a brain tumour in four patients, three of whom had presented with focal seizures and the other with ataxia.
MRI
Following CT head scans, five patients underwent immediate cerebral MRI. In only three patients who had seizures and two presenting with non-seizure events did the results have immediate neurosurgical, neurovascular or neurooncological treatment implications. A further seizure patient had a brain tumour which did not require immediate intervention. MRI scans were subsequently requested, as indicated, by the supervising consultant. In one patient MRI had previously identified mesial temporal sclerosis as the likely cause of epilepsy.
Electroencephalography (EEG)
The results of the 58 EEGs (including seven with video) from the clinic are detailed in Table III. Seventeen (29%) EEGs from the clinic assisted with the diagnosis of epilepsy. An additional four patients had EEGs performed in other departments, prior to first seizure clinic attendance. The classification of epileptic predisposition of 22 EEGs containing epileptiform activity is detailed in Table IV.
Echocardiography:
Three of four patients who underwent echocardiography had newly detected valvular abnormalities; two with aortic stenosis and a third with mitral valve calcification. Of the 112(56%) patients with episodes of altered consciousness attending for 12 lead electrocardiograms (ECGs), four had a borderline or prolonged QTc interval. Cardiology consultants reviewed ECGs with suspected abnormalities. Blood and/or urine tests were performed on 39 patients. Three of four blood alcohol and six of seven plasma gamma glutamyl transferases supported a clinical diagnosis of alcohol dependency. Blood and urine tests were otherwise unhelpful in explaining the reason for attacks. Antiepileptic drug (AED) concentrations supported therapeutic changes made in four of four patients showing signs or symptoms of toxicity.
Outcomes
Fifty patients were invited to review appointments, 42 (84%) of whom attended. Sixty-eight patients with a diagnosis of epilepsy or possible epilepsy were offered further follow up with a neurologist and counselling with an epilepsy nurse specialist. A further 36 patients with other or unclear diagnoses were also offered follow up with a neurologist. Referrals to Cardiology and to the Alcohol and Drug Rehabilitation Services were made either directly or advised to be made locally via the general practitioner. Follow up and referrals to other specialists are detailed in Table V.
AED monotherapy was initiated at the first seizure clinic in 48(24%) patients, 47 for epilepsy and one for an unprovoked first seizure. Twenty-one (10.5%) patients had commenced AED treatment for epilepsy prior to clinic attendance and three for indications unrelated to epilepsy (neuropathic pain or mood stabilization). Six of the 21 individuals had their diagnosis changed from epilepsy to either non epileptic attack disorder (n=2), syncope (n=1), single provoked seizure (n=1), possible first unprovoked seizure (n=1) or uncertain (n=1).
In Scotland there are 20,000 to 40,000 people with active epilepsy and there will be between 2,000 and 3,500 new diagnoses each year.8 During the audit period about one in eight patients seen received a diagnosis of a first unprovoked seizure. Approximately one-third of individuals were diagnosed with epilepsy, thus the remainder either did not have epilepsy or there was insufficient evidence or seizure number to make the diagnosis. First seizure provision should therefore be sufficient to deal with approximately three times the incidence of epilepsy or 150 patients per 100,000 in the adult population per year. This proportion is clearly dependant on referral patterns: where the pool of referring doctors is experienced in the diagnosis of syncope, for example then fewer cases will be referred and the ratio of epilepsy to non epilepsy will increase. In terms of outcome, the importance of the seizure number has been highlighted in long term prospective studies, where the number of seizures prior to commencement of monotherapy in newly diagnosed epileptics was the most important influence on prognosis. The more seizures the worse the prognosis.9 It has been reported that the mean time from first seizure to correct diagnosis in the elderly (mean age 71years) is 19 months,10 whilst in the total population it is 6 months.11 Only 14 (7%) patients overall and two (7.7%) with a first unprovoked seizure were aged over 65 years on first attendance at the seizure clinic. This suggests that the elderly are either not referred, do not attend or tend to present elsewhere. In the older individual a higher likelihood of having unwitnessed events, difficulty in self reporting due to concomitant illness 12and vague symptomatology (with the seizure often being reported as “altered mentation”, “staring” or “unresponsiveness” and the aura as “dizziness”13 ) increases the diagnostic challenge. It has also been suggested that the elderly are more likely to have an extra-temporal lobe focus and are therefore less likely to present with the more easily recognisable classical temporal lobe focal seizure semiology seen in the younger patients.13
Accurate initial assessment and speedy referral of patients presenting with seizure(s) is probably critical to clinical outcome. Excluding five cases with established active epilepsy, only 29% (26 of 89) of individuals in the unprovoked seizure group (new and recurrence from remission) had experienced a single lifetime seizure at initial contact. A further 28% (n = 25) had experienced multiple (>20) unprovoked seizures. There was a male preponderance in the group with seizures related to alcohol and drug withdrawal. It is less clear why there was also a significant excess of males (22 of 26) in those presenting with a single unprovoked seizure. In contrast there was no gender bias in the epilepsy group. Female patients with first seizures may present elsewhere. For instance in pregnancy a first seizure is regarded as eclampsia until proven otherwise and necessitates urgent patient management. Although there were no referrals from maternity services to the first seizure clinic, eclampsia would not account for the majority of first seizures in the female population. Whether females are less likely to be referred or seek help after a single event is uncertain.
An important contributory factor preventing prompt diagnosis in patients is their inability, for whatever reason, to attend outpatient appointments. The combined number of clinic appointment slots cancelled late or not attended was estimated at 35 - 40%. In addition 16% of individuals did not attend follow up first seizure clinic appointments. Whilst seasonal and local holidays contributed to the highest non attendance rates (up to 67%), a socio-economic effect may have been operating in our catchment area. Highest rates of non attendance for clinic appointments are known to be associated with areas of greatest social deprivation.14 Recent attention has focused on the treatment gap and the investigation gap15 and in this audit sub-optimal uptake of services with regards to patient non-attendance would, for a minority of patients, potentially contribute to both. Accurate early diagnosis is also dependent on reliable witness accounts, which may be difficult to obtain. In spite of written requests only three in every five patients were accompanied by a first hand eyewitness. Lack of eyewitnesses at initial contact was one of the main reasons for the need for first seizure clinic review appointments. First hand eyewitness accounts were obtained in 92% (24 of 26) of cases in which a diagnosis of first unprovoked seizure was made, 91%(62 of 68) with epilepsy, but in only 17%(two of 12) in which the diagnosis remained uncertain. Only one patient in the first unprovoked seizure group commenced AED, the decision on treatment being largely decided by the risk of further seizures. Estimates of recurrence rate vary, reflecting differences in selection and methodological criteria. Highest recurrence rates (up to 90%) are seen in patients with epileptiform abnormalities on EEG and an identifiable cause such as congenital neurological deficits16-19 and lowest (13-40%) with a normal EEG and no identifiable seizure aetiology. Overall the risk is30-40%,greatest in the first year after the single event, but less than 10% after two years.20 The risks of third and fourth seizures are much higher than the risk of a second seizure.
Assessing the cost implications and resource utilisation of a first seizure clinic is complex. From the first 200 consecutive attendees, 261 follow up appointments and specialist referrals were generated, 50 at the first seizure clinic for review and 211 with the epilepsy services and elsewhere (excluding referrals to the alcohol rehabilitation and chemical dependency services). The majority (187 of 211, 89%) of follow up referrals were within the Regional Epilepsy Services, to the consultant neurologist, epilepsy nurse specialist, specialist in non epileptic attack disorder, neuropsychologist and to Quarrier’s Epilepsy Assessment Unit.
Follow up is imperative particularly where the diagnosis is uncertain, as recurrent episodes, particularly when witnessed or captured on video EEG, may either enable a more definite diagnosis of epilepsy or refute it. All individuals diagnosed with epilepsy or possible epilepsy at the clinic were offered follow up with both the consultant neurologist specialising in epilepsy and the epilepsy nurse specialist. All patients commencing AEDs had letters sent to their general practitioner with AED protocols, information regarding possible side effects and a copy of DVLA (Driver and Vehicle Licensing Authority) advice given at the clinic. No previously published data for follow up appointments offered and attended are available. However, in the neighbouring region of Lanarkshire, a 1998 survey of patients’ records from 12 general practices, revealed that, in the previous year, no patient receiving anti-epileptic drugs had a recorded review appointment either in primary care or hospital.21 This highlights the need for equality of resources and healthcare throughout Scotland and indeed the UK.
The cost of direct care for epilepsy in the UK has been estimated at £600 million annually with overall costs to the nation of £2 billion, before even considering the cost of lost opportunity to the individual.22 Since this audit, in an attempt to improve uptake of services, promotion of the service to different departments (such as Accident and Emergency) has been initiated.
Identification of factors leading to sub-optimal uptake of services may enable instigation of changes to promote a higher uptake. Implementation of SIGN guidelines,23 by providing fast-track, easy access clinics in acute hospitals (or the new Ambulatory Care and Diagnostic Centres) and prioritising patients with recent onset seizures, should enable their assessment within two weeks. These measures should reduce the number of individuals presenting after multiple events, increase uptake, and thereby improve outcomes.
ACKNOWLEDGEMENTS: I would like to express gratitude to the secretarial staff at the Department of Neuroradiology, the administrative staff at Neurophysiology reception and the medical records staff at the Neurological Institute Outpatient Department for their help. I would also thank the consultant neurologist for advice given during the audit and the cardiologists at the Southern General Hospital for reviewing ECGs.
REFERENCES
1 BK, Cockerell OC, Sander JW et al. The incidence and lifetime prevalence of neurological disorders in a prospective community based study in the UK. Brain 2000; 1123: 665-76.
2 Goodridge DMG, Shorovon SD. Epileptic seizures in a population of 6000:1Demography, diagnosis and classfication, and the role of the hospital services.2. Treatment and prognosis. Br Med J (Clin Res Ed) 1983; 287(6393): 641-47.
3 WA, Kurland LT. The epidemiology of epilepsy in Rochester, Minnesota, 1938 through 1967. Epilepsia 1975; 16: 1-66.
4 King DW, Gallagher BB, Murvin AJ et al. Pseudoseizures: diagnostic evaluation. Neurology 1982; 32: 18-23.
5 Scheepers B, Clough P, Pickles C.The misdiagnosis of epilepsy: findings of a population study. Seizure 1998; 7: 403-06.
6 Smith D, Defalla DA, Chadwick DW. The misdiagnosis of epilepsy and the management of refractory epilepsy in a specialist clinic. Q J Med 1999; 92: 15-23.
7 Scottish Intercollegiate Guideline Network. Diagnosis and management of epilepsy in adults: A national clinical guideline. Edinburgh: 2003; 2.1: 2. and 2.5:1-2
8 Sander JW, Shorovon SD. Epidemiology of the epilepsies. J Neurol Neurosurg Psychiatry 1996; 61: 433-43.
9 Reynolds EH, Heller AJ, Elwes RDC et al.Factors influencing prognosis of newly diagnosed epilepsy. Epilepsia 1989; 30: 648.
10 Spitz MC, Jacquelyn L, Bainbridge K et al. Observations on the delay and diagnosis of seizures in the elderly: update 3. Epilepsia 2003; 44 (suppl 9): 18 (Abstr.1040).
11 Department of Health. Annual Report of the Chief Medical Officer 2001. Chapter 6:Epilepsy-death in the shadows. December 2001. http:// www.doh.gov.uk/cmo/annualreport2001/epilepsy.htm
12 Kramer G. Epilepsy in the elderly: some clinical and pharmacotherapeutic aspects. Epilepsia 2001; 42(suppl 3):55-9
13 Ramsay RE, Pryory F. Epilepsy in the elderly. Neurology 2000; 55:S9-14; discussion S54-8.
14 Can S, Macfarlane T, O’Brien KD. The use of postal reminders to reduce nonattendance at an orthodontic clinic: a randomised controlled trial. Br Dent J 2003; 195(4): 199-201.
15 Kale R. The treatment gap. Epilepsia 2002; 43 (suppl 6): 31-33.
16 van Donselaar CA, Schimsheimer R. Adult patients with untreated idiopathic first seizures. Arch Neurol 1992 Mar; 49(3): 231-37.
17 Hart YM, Sander JW, Johnson AL et al. National General Practice Study of Epilepsy: recurrence after a first seizure. Lancet 1990; 336(8726): 1271-74.
18 Hesdorffer DC, Logroscino G, Cascino G et al. Risk of unprovoked seizure after acute symptomatic seizure: effect of status epilepticus. Ann Neurol 1998; 44: 908-12.
19 van Donselaar CA, Geerts AT, Schimsheimer RJ. Idiopathic first seizure in adult life: who should be treated? Br Med J 1991; 302(6777): 620-23.
20 Berg AT, Shinnar S. The risk of seizure recurrence following a first unprovoked seizure: a quantitative review. Neurology 1991; 41(7): 965-72.
21 Department of Health Annual Report 2001. Section 9.6 Epilepsy. Available from http://www.nhslanarkshire.co.uk/ho/publications/annual_report/epilepsy.htm
22 Bowis J. “Out of the Shadows”: The Political View. Epilepsia 2002; 43 (suppl 6): 16-17.
23 Scottish Intercollegiate Guideline Network. Diagnosis and management of epilepsy inadults: A national clinical guideline. Edinburgh: 2003; 5.2: 29.