Worldwide, approximately 50% of females with epilepsy are of childbearing age.¹ When compared with the general female population, women with epilepsy (WWE) face distinct issues with regard to medication, seizure control, contraception, foetal development, complications of pregnancy, and adverse impact on menopause.^2,3,4 In the UK and Ireland, nine women with epilepsy died during pregnancy or post-partum from 2013-2015.⁵

It is therefore important that healthcare professionals who care for these patients are fully informed and aware of the need for careful planning and management of pregnancy, with the aim of optimising anti-epileptic drug (AED) therapy, minimising maternal and foetal complications, and providing appropriate support before, during, and after pregnancy. 

In particular, the following issues should be considered:

• WWE should receive appropriate counselling to discuss risks, medication and contraceptive implications at least one year prior to conception²
• Certain AEDs are inducers of the hepatic CYP450 system,⁴ and can result in failure of hormonal contraceptive measures, such as hormonal pill, ring or patch. Thus, it is recommended that women taking these medications use appropriate alternative contraceptive methods^1,2
• Folic acid 5mg should be prescribed to all WWE when commencing an AED to reduce risk of foetal neural tube defects⁶
• Foetal exposure to AEDs, especially valproate, increases the risk of malformations.^7,8 Ideally, use of valproate should be avoided in WWE
• WWE on valproate who have an unplanned pregnancy should attend an epilepsy specialist²
• Women who are on AEDs are at higher risk of osteoporosis and should consider calcium supplementation and referral for DXA scan.²

Aims of the audit

To identify women with epilepsy of childbearing age attending an urban GP clinic and, where necessary, optimise management to ensure best practice and compliance with pre-defined standards. 

Standards

The HSE National Epilepsy Care Programme 2014² and European Medicines Agency⁹ guidelines were used to define standards for the purposes of audit.

Methods

The clinic’s electronic patient database was filtered to identify patients meeting the following acceptance criteria: 

• Women aged 16 to 53 prescribed any of the following medications during an 18-month period from January 1, 2014 to June 30, 2015: sodium valproate, phenytoin, phenobarbital, phenobarbitone, carbamazepine, eslicarbazepine, primidone, rufinamide, topiramate, lamotrigine and levetiracetam. 

An extensive list of trade and generic drug name variations was compiled to ensure completeness. Medications known to be CYP450 enzyme inducers were identified. The electronic health record of each patient identified was then accessed and compared with the pre-defined standards. 

Results

A total number of 14 patients were identified. One patient was identified who was not formally listed in the electronic record’s drug-dispensing system as being prescribed an AED (lamotrigine), but reference was made to an active prescription in the consultation notes. This patient was included in the results.

There was poor compliance with national guidelines for management of women with epilepsy. The national guidelines were only published in November 2014 and the period of audit was from January 1, 2014, so the doctors in the practice may not have been aware of the importance of planning pregnancies and of ensuring adequate contraception in women with epilepsy. This highlighted an urgent need to educate the doctors in the practice about the guidelines and to improve the care of women with epilepsy.

Action plan and re-audit

The GPs in the practice held a meeting to discuss the results of the audit and to highlight the guidelines. A re-audit was then planned following implementation of an action plan. 

The clinic’s electronic patient database was once again filtered to identify patients aged 16 to 53 prescribed AEDs during an 18-month period from September 1, 2015 to February 28, 2017.

A total number of 12 patients were identified. One patient was identified who was not formally listed in the electronic record’s drug-dispensing system as being prescribed an AED (lamotrigine), but reference to an active prescription was made in the consultation notes. This patient was included in the results. An additional two patients were identified that did not have a formal diagnosis of epilepsy, but were on an AED for an indication other than epilepsy. These patients were excluded from the results.

Discussion

Coding a diagnosis such as epilepsy is important to improve safety during consultations and to ensure that audits can be carried out quickly. The number of women coded reduced in the re-audit and this should be improved by the doctors in the practice.  

There was a large increase in the percentage of women who had been seen by a neurologist in the previous 18 months, up to 92%. This suggests that almost all WWE were referred to a neurologist for review. At that review the women would have been given information about planning pregnancies. 

Only 50% of women had a documented discussion about contraception in the chart, however since 92% of women had been to the neurology OPD this discussion almost certainly would have occurred there. The number of women on folic acid increased from 29% to 75%, a good improvement but the doctors should aim to achieve the 90% target on this criteria. 

The numbers documented as on contraception or not sexually active is low and needs to be improved. A very low percentage were on calcium or had DXA scan. The risk of osteoporosis in women on AEDs needs to be highlighted to the doctors in the practice again. 

There was an increase in the percentage of women on valproate. Since a high percentage of women had attended the neurologist for review it is most likely that all women were advised to change from valproate to an alternative and it is possible that some opted to stay on valproate and were either not planning a pregnancy and either on effective contraception or not sexually active. It would be important to reiterate the risks of valproate at every visit to these women.

Challenges

It was necessary to use AED prescriptions as a means of identifying patients. For the purpose of safety, this ensured that WWE were not missed regardless of whether their epilepsy diagnoses had been appropriately coded. It is noteworthy that this approach was technically challenging. Firstly, it required an exhaustive list of all relevant drugs, including generic names, trade names, and formulations. This task was carried out by hand using a combination of MIMS and the practice management software’s drug formulary. Once the list had been created, it could then be used to query the electronic patient database. 

However, this revealed certain limitations of the practice management software. In particular, it was not possible to achieve the desired filter using the software’s built-in ‘custom reports’ feature. It appears that this was due to an unspecified limit on the number of drugs that could be entered simultaneously when building a custom report. After contacting technical support, it was ultimately necessary to query the database directly, and to break the full drug list into a number of smaller queries that could be run one after another. Ultimately, this allowed all relevant patients to be identified. 

This highlights the fact that a certain degree of technical expertise was necessary to complete what should have been a relatively simple process. Furthermore, if another  practice wished to carry out the same audit, it is not immediately clear how these database queries could be easily shared. 

Arguably, there would be value in practice management software incorporating a feature to allow such ‘custom’ audit queries to be made accessible to other practices, thereby allowing other GPs to easily perform the same audit. This would help to streamline and standardise the audit process, and ultimately could help generate deeper insights from the data generated at a national level. 

References

1. Gooneratne IK, Wimalaratna M, Ranaweera AKP, Wimalaratna S. Contraception advice for women with epilepsy. BMJ. 2017; 357
2. HSE. Effective management of Women with Epilepsy in the non-acute setting: Ambulatory SOP Development Workgroup; 2014 [Available from: https://www.icgp.ie/go/courses/women_s_health/programme_news/2A9FD029-F220-F32B-721AB5D6D28D449B.html
3. MacDonald SC, Bateman BT, McElrath TF, Hernandez-Diaz S. Mortality and Morbidity During Delivery Hospitalization Among Pregnant Women With Epilepsy in the United States. JAMA neurology. 2015; 72(9): 981-8
4. O’Brien M, Gilmour-White S. Management of epilepsy in women. Postgraduate Medical Journal. 2005; 81(955): 278-85
5. Knight M NM, Tuffnell D, Shakespeare J, Kenyon S, Kurinczuk JJ. Saving Lives, Improving Mothers’ Care: Lessons learned to inform maternity care from the UK and Ireland Confidential Enquiries into Maternal Deaths and Morbidity 2013–15. MBRRACE-UK; 2017
6. Kaaja E, Kaaja R, Hiilesmaa V. Major malformations in offspring of women with epilepsy. Neurology. 2003; 60(4): 575-9
7. Holmes  LB, Harvey  EA, Coull  BA, Huntington  KB, Khoshbin  S, Hayes  AM, et al. The Teratogenicity of Anticonvulsant Drugs. New England Journal of Medicine. 2001; 344(15): 1132-8
8. Artama M, Auvinen A, Raudaskoski T, Isojarvi I, Isojarvi J. Antiepileptic drug use of women with epilepsy and congenital malformations in offspring. Neurology. 2005; 64(11): 1874-8
9. PRAC recommends strengthening the restrictions on the use of valproate in women and girls [press release]. European Medicines Agency, October 10th 2014 2014