If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Despite NICE guidance, visits to neurosurgery clinics remain uncommon and are seldom followed by surgical interventions.
•
Access to tertiary centres and neurosurgical procedures in adults with epilepsy in England, varies by region and CCGs.
•
Access to neurosurgical procedures is delayed, particularly if a person’s initial visit is not at a tertiary epilepsy centre.
•
VNS use remains low, although it is higher in people with learning disability, who are less likely to have resective surgery.
Abstract
Purpose
To understand if primary consultation at tertiary epilepsy centres (TEC) in England impacts access to neurosurgical procedures (resective surgery, vagus nerve stimulator [VNS], deep brain stimulator [DBS]).
Methods
Adults with epilepsy, and with a first neurology outpatient visit (index) between 01/01/2013 and 31/12/2015, were followed using English Hospital Episode Statistics from index date to 31/12/2019. Analyses were stratified by geographic location, learning disability record, and whether the index or follow-up visits were at a TEC.
Results
84,093 people were included, with mean 5.5 years of follow-up. 12.4% of the cohort had learning disability (range 10.1%-17.4% across regions). TEC consultations varied by National Health Service regions and Clinical Commissioning Groups. 37.5% of people (11.2%-75.0% across regions) had their index visit at a TEC; and, of those not initially seen at a TEC, 10.6% (6.5%-17.7%) subsequently attended a tertiary centre. During follow-up, 11.1% people (9.5%-13.2%) visited a neurosurgery department, and 2.3% of those (0.9%-5.0%) then underwent a neurosurgical procedure, mainly VNS implantation. Median time from index date to first visit at a neurosurgery centre was 7 months (range 6-8 months across regions) and 40 months to procedure (36.5-49 months, 37.0 months in people with index visit at a TEC and 49.0 months otherwise). People with learning disability were less likely to have resective surgery (<0.5% versus 1.0% in those without) and more likely to undergo VNS implantation (5.8% versus 0.8%).
Conclusion
Although clinically recommended for suitable individuals, neurosurgical procedures in epilepsy remain uncommon even after consultation at a TEC. Geographical variation in access to TECs was present.
Despite numerous pharmacological options for people with epilepsy, the goal of seizure freedom without side-effect burden remains elusive for a large proportion of individuals. Treatment with anti-seizure medications (ASMs) typically starts with one ASM, whereafter the regimen is adapted with changes to drug and dose until seizures are controlled and the treatment is tolerated [
People who do not attain seizure freedom with the first or second ASM have markedly reduced chances of successful pharmacological treatment with subsequent new medications and increased likelihood of developing drug-resistant epilepsy (DRE) [
Treatment outcomes in patients with newly diagnosed epilepsy treated with established and new antiepileptic drugs: a 30-year longitudinal cohort study.
]. DRE is defined by the International League Against Epilepsy (ILAE) as “as failure of adequate trials of two tolerated, appropriately chosen and used antiepileptic drug schedules (whether as monotherapies or in combination) to achieve sustained seizure freedom”. DRE is associated with serious consequences, including neuropsychological decline, decreased quality of life, and increased risk of death [
]. The prevalence of DRE is estimated as approximately 30% (95% confidence interval between 19% and 42%), a proportion which has remained relatively constant over the past decades [
Treatment outcomes in patients with newly diagnosed epilepsy treated with established and new antiepileptic drugs: a 30-year longitudinal cohort study.
In people with DRE, several non-pharmacological options exist. These include resective surgery, corpus callosotomy, and the implantation of neuromodulation devices (vagus nerve stimulator [VNS] or deep brain stimulator [DBS]) and dietary therapy [
]. Approximately two-thirds of people with intractable temporal lobe epilepsy, and half of people with focal neocortical epilepsy, have been shown to achieve seizure freedom following resective surgery [
]. For VNS, 45%-65% of individuals demonstrate at least a halving and, in some cases up to 100% reduction, in seizure frequency within 6 months of initiating treatment [
]. The 2012 National Institute for Health and Care Excellence (NICE) clinical guideline 137 (applicable at the time of data capture for this study) recommended prompt referral to tertiary services for people with uncontrolled epilepsy to assess suitability for resective surgery or alternatives, and the recently updated guidance 2022 further strengthens support for this approach [
]. The NICE guidelines recommend considering VNS as an adjunctive treatment for people who are refractory to ASMs and are not suitable for resective surgery. This includes people whose epilepsy disorder is dominated by focal seizures (with or without secondary generalisation) or generalised seizures [
An alternative neuromodulatory approach is DBS implantation, which has been available in England since 2012. Deep brain simulation has been reviewed under NICE interventional procedures guidance [
]. Anterior thalamic stimulation targets can be considered in people with a diagnosis of refractory epilepsy and under specific arrangements, although routine use of DBS for epilepsy is not commissioned by the National Health Service (NHS) England Specialised Services Commissioning [
Despite NICE clinical guidance on the management of people with DRE, though, access to non-pharmacological treatment options is limited and is seldom considered early in the treatment pathway [
Samanta D., Ostendorf A.P., Willis E., Singh R., Gedela S., Arya R., et al. Underutilization of epilepsy surgery: Part I: a scoping review of barriers. Epilepsy Behav 2021;117:107837. Apr.
]. Given the adverse consequences of uncontrolled epilepsy, variation in referral and access to potential non-pharmacological options for people with DRE may have a detrimental effect on clinical outcomes, and increase the burden on the healthcare system.
In this context, we investigated referral to neurology and neurosurgery departments for adults wih epilepsy in England, and access to epilepsy neurosurgical procedures including resective surgery, VNS and DBS, depending on people's pathway of care either via a tertiary epilepsy centre (TEC) or a non-specialist centre (NSC). We hypothesised that the causes of variation in access may include the organisational structure of the NHS at a regional and Clinical Care Commissioning Group (CCG) level, differences in clinical practice or individual clinical characteristics for example, the presence of a learning disability.
]. The HES database contains records of NHS-funded and private-funded people treated by an NHS provider and the records of non-English residents treated in hospital settings. The database also contains the records of admissions to independent (non-NHS) providers where treatment is funded by the NHS. Disease diagnoses in HES are recorded using the International Classification of Diseases 10th revision (ICD-10) codes; and interventions, procedures and procurement for treatment are recorded using the Office of Population Censuses and Surveys Classification of Interventions and Procedures version 4 (OPCS-4) codes. Our study observation period included all HES records dated between 31/12/2011 and 31/12/2019.
The study population comprised people with both (i) a visit to a neurology outpatient department (treatment speciality code 400, the first ever neurology visit during the observation period and flagged as a “first visit”) at ≥18 years of age between 01/01/2013 and 31/12/2015 inclusive (index date); and (ii) a recorded diagnosis of epilepsy (ICD10 codes G400-G419) anywhere in their HES records during the study observation period. Code lists are provided as supplementary material. The first outpatient visit to a neurosurgery department post index date was also extracted to define attendance at neurosurgery. People with a record of brain tumour (ICD-10 codes C710-C718) on or prior to the neurosurgery first visit were excluded. The hospital visits may have been to see a neurologist, a neurosurgeon or other relevant clinical expert(s), for example, a multidisciplinary team, to assess the possibility of neurosurgical procedures, to conduct investigations, or to perform a procedure. All people were followed from index date to the end of study period on 31/12/2019, i.e., before the impact of the COVID-19 pandemic affected NHS practice.
2.2 Variable definitions
Procedures and investigations extracted included those performed at TECs and NSCs. Procedures comprised a first code for either resective surgery, VNS or DBS implantation specifically performed for a diagnosis of epilepsy, and recorded during an inpatient admission in the period between an individual's first neurosurgery visit and the end of follow-up. Investigations were extracted in people undergoing a procedure from the index date up to the date of the first procedure and comprised telemetric video electroencephalogram (vEEG) during an elective admission to either neurology or neurosurgery (speciality code 150 or 400), and, as an inpatient or out-patient at neurology or neurosurgery, computed tomography (CT) of the head or magnetic resonance imaging (MRI).
To investigate geographic variation at a regional level in England, analyses were stratified by seven geographic regions based on the NHS Region of residence of people at index date: East of England, London, Midlands, North East and Yorkshire, North West, South East, and South West and further sub-divided by CCG.
No published list of TECs in England was available; therefore a TEC was defined as a hospital with both neurology and neurosurgery departments that could offer outpatient and inpatient care, in accordance with NHS England Specialised Services Commissioning requirements for epilepsy neurosurgical procedures [
]. TECs were assigned to an NHS region based on the geographical location of the centre. A TEC visit was defined as a neurology or neurosurgical visit at a centre identified as being a TEC.
Individuals were identified as having a learning disability if they attended a learning disability clinic (treatment function code 700) or had a recorded diagnosis of learning disability (see code list in Appendix) at any time in their outpatient or inpatient records during the study observation period.
2.3 Data analysis
All analyses were performed (i) for the total study population, and stratified by (ii) NHS region; (iii) individual's history of a learning disability; and (iv) index visit at a TEC. To evaluate geographical variation more precisely, the percentage of people attending a TEC either at index date or during follow-up was also estimated by CCGs. Categorical variables were summarised using frequency counts and percentages, overall and specifically in those attending neurosurgery, whenever relevant. Continuous variables were summarised using counts, arithmetic mean and standard deviation (SD) if normal, or median and interquartile range (IQR) if non-normal (based on Kolmogorov-Smirnov tests). Visits to a neurosurgical department, and procedures were also described in terms of time to access. Categorical comparisons were performed using Chi-square tests. Medians for time-to variables were compared across strata using either the Mann-Whitney U-test for binary comparisons of TEC versus NSC, and comparing comparing people with versus without learning disability; or the Kruskal-Wallis test for tests of differene across regions (as distributions were not normal).
3. Results
The study population comprised 84,093 people (Supplementary Fig. 1) with a mean age of 49 years (SD: 20) and mean follow-up of 5.5 years (SD: 0.9). A subset of 1,988 (2.4%) people did not have geographical information available and were excluded from analyses by NHS region and CCGs. Demographic characteristics; and descriptions of access to TEC, neurosurgery departments, and procedures, are described by NHS region in Table 1.
Table 1Access to adult epilepsy services after a first visit to neurology between 2013 and 2015, overall and stratified by NHS Region.
England region
England1
East of England
London
Midlands
North East & Yorkshire
North West
South East
South West
p-value
Variable
84,093
9152
11,430
15,829
14,041
12,884
11,332
7,437
Follow-up post-index date, years
5.5 (0.9)
5.5 (0.9)
5.6 (0.9)
5.5 (0.9)
5.6 (0.9)
5.4 (0.9)
5.5 (0.9)
5.5 (0.9)
Descriptive characteristics
Age, mean (SD)
49 (20)
51 (20)
49 (20)
50 (20)
49 (19)
49 (19)
50 (20)
50 (20)
Female, %
49.7%
49.7%
50.2%
50.6%
49.2%
48.8%
50.4%
48.9%
Learning disability, %
12.4%
11.0%
10.1%
11.4%
17.4%,
11.8%
12.5%
12.6%
Attendance at a TEC:
For an outpatient neurology visit at index, % (N)
37.5% (31,508)
11.2% (1,029)
28.3% (3,232)
24.5% (3,874)
57.0% (8,009)
75.0% (9,662)
25.9% (2,937)
21.5% (1,599)
<0.0001
For an outpatient neurology visit at index or follow-up, % (N)
44.1% (37,060)
18.6% (1,701)
38.0% (4,342)
32.7% (5,172)
62.7% (8,808)
79.4% (10,233)
31.4% (3,561)
26.6% (1,977)
<0.0001
Outpatient visit a neurosurgery department at index or follow-up, % (N)
11.1% (9,347)
9.5% (873)
12.3% (1,404)
10.7% (1,689)
11.3% (1,587)
13.2% (1,706)
9.8% (1,110)
10.2% (760)
<0.0001
Outpatient visit a neurosurgery department at index or follow-up, % (N) in those ever attending a TEC
14.3% (5,313)
19.5% (331)
16.7% (726)
16.0% (825)
13.0% (1,141)
13.1% (1,344)
12.8% (457)
16.5% (327)
<0.0001
Time from index neurology visit to first neurosurgery visit (months), median (IQR)
7 (2-27)
6 (2-24)
6 (2-27)
6 (2-24)
7 (2-26)
8 (2-29)
7. (2-26)
8 (2-30)
0.001
Epilepsy procedures in people seen at neurosurgery department, N (%)
At least one procedure
212 (2.3%)
8 (0.92%)
20 (1.4%)
42 (2.5%)
34 (21%)
26 (1.5%)
30 (2.7%)
38 (5.0%)
<0.0001
Resective surgery2
85 (0.91%)
<5
12 (0.85%)
8 (0.47%)
13 (0.82%)
15 (0.878%)
13 (1.2%)
8 (1.1%)
0.292
VNS2
121 (1.3%)
<5
7 (0.50%)
31 (1.8%)
20 (1.3%)
12 (0.70%)
15 (1.4%)
29 (3.8%)
<0.0001
DBS2
10 (0.11%)
<5
<5
<5
<5
0
<5
<5
NA3
Time from index neurology visit to first procedure (months), median (IQR)
40 (27-53)
48.5 (40-62)
49 (35-63)
38 (26-50)
39 (27-53)
36.5 (27-53)
45.5 (28-51)
47 (27-54)
0.161
1The England column includes 1988 people with missing region of residence. 2Numbers when fewer than 5 people in a cell are suppressed; a p-value is not presented for DBS due to the small cell counts. DBS, deep brain stimulation; IQR, interquartile range; N, number of people; NA: Not available; SD, standard deviation; TEC, tertiary epilepsy centre; VNS, vagus nerve stimulation.
The index visit was at a TEC in 37.5% of individuals, with wide variation across regions ranging from 11.2% in East of England to 75.0% in the North West (Table 1). Overall, 44.1% of people visited a TEC either at or post the index date, varying by region (18.6%-79.4%) and across CCGs (Fig. 1). Of people first seen at a TEC, almost all had a further visit at a TEC during follow-up (90.4%, range across regions: 88.6%-94.4%), whilst people seen at an NSC at index were unlikely to subsequently attend a TEC during follow-up (10.6%, range: 6.5%-17.7%). The proportion of the study population seen in neurosurgical departments was consistently higher for people who ever attended a TEC at index (13.5% versus 9.7%) across NHS regions, except in the North West region (Table 2 and Supplementary Table S1).
Fig. 1Visit at a TEC at or post-index date by Clinical Commissioning Groups in England
Percentage of people attending a TEC post-index date in each CCG, with overlay of NHS regions. Each CCG is colour-coded according to the percentage of people with epilepsy attending a TEC.
Table 2Access to adult epilepsy services after a first visit to neurology between 2013 and 2015, overall and stratified by NHS Region and by access to a tertiary epilepsy centre.
First neurology visit
Variable
TEC, N (%)
NSC, N (%)
p-value
Number of people, %
31,508 (37.5%)
52,585 (62.5%)
Attendance at a TEC during follow-up, %
90.4%
10.6%
<0.0001
Attendance at neurosurgery department during follow-up, %
13.5%
9.7%
<0.0001
In people attending neurosurgery department, N
4240
5107
At least one procedure, %
3.1%
1.6%
<0.0001
VNS, %
1.7%
1.0%
<0.0001
Resective surgery, %
1.3%
0.55%
<0.0001
DBS, %
0.1%
<5
Time from index neurology visit to first neurosurgery visit (months), median (IQR)
7 (2-26)
7 (2-27)
0.10
Time from index neurology visit to first procedure (months), median (IQR)
37 (24-50)
49 (34-58)
<0.0001
DBS, deep brain stimulation; IQR, interquartile range; N, number of individuals; NSC, non-specialist centre; SD, standard deviation; TEC, tertiary epilepsy centre; VNS, vagus nerve stimulation.
In total, 11.1% of people visited a neurosurgery department during follow-up (range 9.5% to 13.2% by region), and this rose to 14.3% (12.8%-19.5%) in people attending a TEC at index or post-index visit. Among people who made at least one visit to a neurosurgery department, 2.3% underwent a procedure during follow-up (0.25% overall): 1.3% for VNS, 0.91% for resective surgery and 0.11% for DBS (Table 1). The median time for people from index visit at any neurology department to first visit at a neurosurgery department was 7 months (IQR: 2-27 months), and was similar in people attending and not attending a TEC at index date. The median time from index date to undergoing a surgical procedure was 40 months (IQR: 27-53 months). The shortest average times from index date to surgery were in the North West (36.5 months), and the longest were in London (49 months) and East of England (48.5 months). Those with a first visit at a TEC had both a higher chance of having a neurosurgical procedure (3.1% if seen at a TEC versus 1.6% if the index visit was at an NSC), and the median time to surgery was on average 1-year shorter (37 months versus 49 months). For VNS specifically, the proportion undergoing implantation was 1.7% in those with index visit at a TEC and 1.0% in those with index visit at an NSC; and for resective surgery the proportions were 1.3% and 0.55%.
In people having VNS implantation during follow-up, 48.5% had a prior record of vEEG, 41.9% had an MRI and 11.0% had a CT. In those undergoing resective surgery, 45.8% of people had a recorded vEEG investigation, 49.0% of people had a MRI in 49.0% , and 50.0% of people had a CT (Supplementary Table S2).
In our study, 12.4% of participants (10.1%-17.4% by regions) had learning disability (Table 3) , see Supplementary Table S3 for regional stratification). Among people with learning disability, attendance at a TEC during follow-up was slightly higher than in people without a record of learning disability (46.9% versus 43.7%). By contrast, attendance at a neurosurgery department was lower in people with learning disability (9.1% versus 11.4% in people without). People with learning disability were more likely to undergo VNS implantation than people without a learning disability (5.8% versus 0.79% of those with a neurosurgical appointment; 0.53% versus 0.090% in the entire study population) and less likely to undergo resective surgery (0.52% versus 0.95% of those seen in neurosurgery; 0.048% versus 0.11% out of the entire study population) compared to people without learning disability, although absolute numbers remained low. The patterns of regional variation when stratified by history of learning disability were similar to that observed in the overall study population.
Table 3Access to adult epilepsy services after a first visit to neurology between 2013 and 2015, overall and stratified by NHS Region and by people’s history of learning disability.
Learning disability
Variable
Yes
No
P-value
Number of people, %
10,456 (12.4%)
73,637 (87.6%)
Attendance at a TEC during follow-up (%)
46.9%
43.7%
<0.0001
Attendance at a neurosurgery department (%)
9.1%
11.4%
<0.0001
In people attending neurosurgery department, N
955
8,392
At least one procedure, %
6.6%
1.8%
<0.0001
Resective surgery, %
0.52%
1.0%
<0.0001
VNS, %
5.8%
0.8%
<0.0001
DBS, %
0.31%
0.083%
NA
Time from index neurology visit to first neurosurgery visit (months) Median (IQR)
11 (3-33)
6 (2-26)
<0.001
Time from index neurology visit to first procedure (months) Median (IQR)
39 (21-51)
40 (28-54)
0.12
Numbers with fewer than 5 people in a cell are suppressed and a Chi-square result is not presented for these variables due to the small cell counts. DBS, deep brain stimulation; IQR, interquartile range; N, number of people; SD, standard deviation; TEC, specialist epilepsy centre; VNS, vagus nerve stimulation; % N: percentage of the study population; % neurosurgery: percentage of people restricting denominator to population with a neurosurgery visit during follow-up.
This large study of all adults in England with epilepsy and a first neurology visit between 01/01/2013 and 31/12/2015, who had on average 5.5 years of follow-up, demonstrates large regional variations in access to a TEC, referral to neurosurgery and rates of VNS implantation.
Our study findings highlight the low rates of adoption of neurosurgical options for people with epilepsy in England within 4-7 years from diagnosis. Within regions and overall, the percentage of people undergoing either resective surgery, VNS or DBS was very small (0.25% overall).
Overall 44.1% of people visited a TEC at index or during follow-up, although we observed large NHS regional and CCG-level variations, indicating a lack of uniformity in pathways of care acros England. Over a third of individuals had their index visit at a TEC neurology department. Only a small minority not seen at a TEC at their index visit were later referred to a TEC (10.6%). According to NICE clinical guidance and existing evidence, people who develop DRE should be prioritised for referral to a TEC, therefore our findings may indicate that access to a TEC is sometimes governed by local NHS referral pathways rather than clinical need [
]. People attending a TEC at index were more likely to attend a neurosurgery department during follow-up, although there was regional variation, and their time from index visit at the neurology department to neurosurgery was around a year shorter.
Individuals attending a TEC at index visit had similar rates of neurosurgical procedures to those who started their journey at a NSC, showing that TEC access was not a strong determinant of undergoing a neurosurgical procedure in our study. Furthermore, regional variation in TEC attendance did not visibly correlate with the proportion of those undergoing neurosurgical procedures in a region.
Only 11% of people with epilepsy in our cohort had an appointment at a neurosurgery department during the follow-up period, which is lower than may be expected assuming a prevalence of DRE of approximately 30% in adults in England [
]. Fewer than 3% of adults with epilepsy seen in a neurosurgery clinic underwent an operation during follow-up, a finding consistent across regions. It is possible that this low percentage is partially due to the relatively short follow-up of our study with a median follow-up of 5.5 years (IQR: 0.9), with a median time from index appointment to surgical procedure of over 3.3 years (40 months). The start of our study period was chosen to correspond to the publication of the NHS England Commissioning Policy on VNS in 2013 [
]; and the end was determined to be prior to the onset of the COVID-19 pandemic which substantially altered access to healthcare. Previously, it has been reported that the average time from first epilepsy diagnosis to resective surgery can exceed 20 years in some people; although the best outcomes are achieved when surgery is performed within 5 years after diagnosis [
]. Therefore our observation of very low rates of resective surgery during our study period is perhaps indicative of a lack of provision at a time when individuals are likely to benefit most from the procedure. The causes of the treatment gap that arise with delayed neurosurgical referrals have been previously investigated and shown to include an array of factors including a lack of knowledge among physicians, inequitable access to specialist epilepsy centres and the inherent complexity of pre-surgical investigations. Biases, either conscious or unconscious, may also influence choosing surgical candidates, for example socioeconomic characteristics or presence of a learning disability [
Samanta D., Ostendorf A.P., Willis E., Singh R., Gedela S., Arya R., et al. Underutilization of epilepsy surgery: Part I: a scoping review of barriers. Epilepsy Behav 2021;117:107837. Apr.
]. Importantly, people with shorter waits to surgical assessment have been shown to have more favourable seizure outcomes, while early diagnosis and access to specialist neurology services and treatments are significant factors in achieving seizure control [
Fahoum F., Boffini M., Kann L., Faini S., Gordon C., Tzadok M., et al. VNS parameters for clinical response in Epilepsy. Brain Stimul. 2022;15(3):814–821. May.
] although eligibility for VNS is likely to be higher given its broader indications for use in people with DRE. The proportion of people undergoing surgical intervention in our study (0.25% of the overall cohort) remains very small compared to the total epilepsy population; and hence relative to the expected proportion of people with DRE who may be eligible for non-pharmacological options. This may indicate either (i) inappropriate levels of referral to neurosurgery, (ii) a long delay for some individuals (beyond the 7 years of our sudy follow-up), and/or (iii) a lack of access to resective surgery or neuromodulation options.
Previous analyses of adult epilepsy procedures across the UK reported low and decreasing rates in resective surgery from 422 to 242 between 2000 and 2011, with an increase in VNS implantation over the same period from 159 to 226 procedures [
]. In our study population which included people with epilepsy first visiting a neurology department in 2013-2015, 121 people were implanted with VNS between 2013 and 2019, and 85 underwent resective surgery, representing 1.43 people per 1000 for VNS and 1.04 per 1,000 for resective surgery.
A decline in surgical procedures has been reported elsewhere in Europe and North America [
]. A drop in the rates of surgical interventions might be surprising given the published evidence on the superiority of resective and non-resective epilepsy surgery procedures over continued medical treatment alone, for people with DRE [
Neligan A., Haliasos N., Pettorini B., Harkness W.F., Solomon J.K.. A survey of adult and pediatric epilepsy surgery in the United Kingdom. Epilepsia 2013;54(5):e62-e65.
VNS was more frequently implanted in people with learning disability than in those without, although this may be expected as people with learning disability may be less likely to be candidates for resective surgery, and, the use of VNS was low even in this population. Very few people with an underlying learning disability had resective epilepsy surgery, and absolute numbers were lower than expected based on cohort size. Importantly, the recently updated NICE guideline states that people with genetic abnormalities or learning disability should not be excluded from referral to a tertiary centre for the consideration of surgery [
], and this may lead in the future to an increase in people with learning disability being referred for neurosurgical assessment.
The strength of this study is that it provides large-scale real-world evidence about the clinical management of people with epilepsy across England.. Due to its coverage, access to HES provides accesss to comprehensive sample of people living in England and diagnosed with epilepsy, consequently, the study results may inform public health decisions at a national level.
Given the reported prevalence of DRE and its associated mortality and morbidity, the reasons for low rates of surgical intervention both in people seen at NSCs or at a TEC, and for patterns of regional variation in access to neurosurgery referrals and procedures, require further investigation. For this, an assessment of how many patients have DRE would be needed, but that is outside the scope of this research study.
The study is subject to limitations.
The identification of a TEC was made assuming that hospitals with both neurology and neurosurgery departments that could offer outpatient and inpatient care provided TEC services to people with epilepsy, however misclassification may have occured if some hospitals with both services were not organised as a TEC. A visit to neurology or neurosurgery departments was assumed to be for epilepsy, as the study population has a history of epilepsy and procedures had a linked epilepsy diagnosis code; however, some individuals may have been referred for other reasons such as diagnostic uncertainty, access to neuropsychological assessment or another specialist setting. Similarly, we were not able to ascertain the reason for attendance at a TEC. An identification of DRE status was out-of-scope in this study as its determination would require additional data linkage to a primary care dataset to review numbers of medications prescribed over the duration of the study.
The average delay between diagnosis of DRE and epilepsy surgery has been previously reported consistently across countries at approximately 20 years in adults and 5 years in children [
]. Our study follow-up ranged 4 to 7 years, therefore, it may have been insufficient to identify all indivduals undertaking neurosurgical procedures after being referred to neurology departments and thoseidentified may have had more severe disease or other characteristics not captured in this study. Follow-up ceased in 2019 to avoid the COVID-19 pandemic influencing results, which means this study may not capture most recent clinical practice. Given the impact of COVID-19 on the healthcare of people with epilepsy in the England, though, the pandemic is likely to have resulted in fewer procedures and longer delays to operative assessment [
As this study is based on electronic health records, it is restricted to those data that are routinely recorded. The low proportion of people with investigations prior to surgery may be due to under-recording and warrants further exploration. While the HES data can follow people across English NHS settings, those who leave the English NHS (due to relocation or death) are not always identified, so may continue to contribute to follow-up time. There may have been under-recording of learning disability if the diagnosis was made in primary care rather than secondary care, and therefore was not recorded in HES; however both clinical codes and referral to services were searched to minimise under-ascertainment. Finally, the study is predominantly descriptive and did not directly compare variables with adjustment for potential confounders.
5. Conclusion
In conclusion, rates of access to TEC vary across the country; which warrants further investigation. Assessment for surgical interventions are recommended for people with DRE, which is estimated to affect as many as 30% of people with epilepsy. Our data show that a very small minority of people with epilepsy undergo a surgical intervention within the first 4-7 years after their first visit to a neurologist, suggesting a large treatment gap and delays to clinically effective options.
Funding
The work was supported by LivaNova PLC (registered in England and Wales with registered no. 09451374), with its registered office address at 20 Eastbourne Terrace - London - W2 6LG.
Highlights
1.
Despite NICE guidance, visits to neurosurgery clinics remain uncommon and are seldom followed by surgical interventions.
2.
Access to tertiary centres and neurosurgical procedures in adults with epilepsy in England, varies by region and CCGs.
3.
Access to neurosurgical procedures is delayed, particularly if a person’s initial visit is not at a tertiary epilepsy centre.
4.
VNS use remains low, although it is higher in people with learning disability, who are less likely to have resective surgery.
Declaration of Competing Interest
JM, FB, VD and MD are employees of LivaNova PLC. SB and JW are employees of Harvey Walsh Ltd., which is part of OPEN Health Ltd, and MA is employee of OPEN Health Ltd. GCH is employed by Gillian Hall Epidemiology Ltd which received funding from OPEN Health Ltd. for scientific input into the study. AS is supported by the Oxford NIHR Biomedical Research Centre.
Treatment outcomes in patients with newly diagnosed epilepsy treated with established and new antiepileptic drugs: a 30-year longitudinal cohort study.
Samanta D., Ostendorf A.P., Willis E., Singh R., Gedela S., Arya R., et al. Underutilization of epilepsy surgery: Part I: a scoping review of barriers. Epilepsy Behav 2021;117:107837. Apr.
Fahoum F., Boffini M., Kann L., Faini S., Gordon C., Tzadok M., et al. VNS parameters for clinical response in Epilepsy. Brain Stimul. 2022;15(3):814–821. May.
Neligan A., Haliasos N., Pettorini B., Harkness W.F., Solomon J.K.. A survey of adult and pediatric epilepsy surgery in the United Kingdom. Epilepsia 2013;54(5):e62-e65.
00282-5&id=gr1.jpg){kind=link}