NIHR Signal Prescribing regular drugs to prevent febrile convulsions risks more harm than benefit
Published on 1 August 2017
The benefits of giving anti-epileptic medication to children aged between six months and seven years who have had a convulsion while feverish, do not outweigh the harms. While diazepam given when a child becomes feverish reduced the chance of a convulsion from about 25% to 18% after a year, one in three children experienced adverse effects. Therefore, the authors suggest benefits do not seem to outweigh the harms.
In the UK 2 to 5% of children less than six years old will have a febrile convulsion, with around a third experiencing another when feverish. Febrile convulsions are usually brief and harmless, though they are disturbing to witness and cause worry for parents and carers. Efforts to reduce the child’s temperature, if they are distressed by a fever are still encouraged.
NICE does not currently recommend preventative medication for febrile convulsions based on the balance of benefits and harms. These systematic review findings support this. Parents and families should be supported with advice and information about how to manage future convulsions and the benign nature of the condition.
- Child Health, Infections, Medicines
Why was this study needed?
Febrile convulsions are fits that can occur when a young child (aged six months to six years) has a fever. Around 2 to 4% of children will have a febrile convulsion, with a third of those children going on to have another when feverish in the future. Most convulsions last under five minutes and the child may feel sleepy for a while afterwards.
Simple febrile convulsions, by definition, are harmless and not a sign of serious brain disease, so can be managed with basic first aid such as putting the child in the recovery position and reducing the child’s temperature if they are distressed. However, ensuring that there is not another cause of a convulsion may require a hospital assessment, especially in children under 18 months old.
For children thought to be at particular risk of recurrence it is tempting to prescribe drugs to try to prevent further fits. However, this approach is unproven and the potential side effects of treatment may outweigh any benefits. This review aimed to gather the evidence on the effectiveness and safety of preventative drugs.
What did this study do?
This systematic review and meta-analysis included 30 randomised or quasi-randomised trials that compared drugs given continuously or intermittently to prevent febrile convulsions with each other, placebo or no treatment in 4,256 children.
The included trials evaluated the anti-epileptic drugs phenobarbitone, phenytoin and valproate; benzodiazepines diazepam and clobazam (sedatives); antipyretics (to reduce fever) paracetamol, ibuprofen and diclofenac; and dietary supplements pyridoxine (vitamin B6) and zinc sulphate.
The majority of included studies were 20 to 30 years old and generally moderate to poor quality. Common sources of bias were lack of blinding; the possibility that patient characteristics influenced group allocation; and incomplete outcome-reporting for all participants. The analysis also suggested some publication bias, meaning studies that found an effect of treatment were more likely to be published than those that didn’t.
What did it find?
- Intermittent diazepam reduced risk of another febrile convulsion at 12 months (relative risk [RR] 0.69, 95% confidence interval [CI] CI 0.56 to 0.84; eight studies, n=1,416), and 24 months (RR 0.73, 95% CI 0.56 to 0.95; four studies, n=739) compared with placebo or no treatment.
- By six months 11% of children given diazepam had a recurrent convulsion compared with 18% in the control group, with 16 needing treatment to prevent one convulsion. By 12 months the convulsion rate was 18% vs. 25% with 13 needed to treat.
- Continuous phenobarbitone reduced the risk of a convulsion at 12 months (RR 0.54, 95% CI 0.42 to 0.70; seven studies; n=807) and 24 months (RR 0.69, 95% CI 0.53 to 0.89; three studies, n=533) compared with placebo or no treatment, but not at 18 or 72 months. The number needed to treat to prevent one convulsion was 14 at six months and eight at 12 months.
- Intermittent clobazam reduced the risk of another fit at six months (RR 0.36, 95% CI 0.20 to 0.64). However, this single study (n=30) also reported a very high rate of fits in children receiving placebo or no treatment (83%), so it is not clear whether these results are reliable.
- The following medications had no statistically significant effect on convulsion recurrence compared with placebo or no treatment: intermittent phenobarbitone, phenytoin, valproate, intermittent clobazam, intermittent ibuprofen, pyridoxine (vitamin B6) or zinc sulphate. There were also no significant findings in studies assessing drugs in combination or comparing them with each other.
- Adverse effects were variably reported across studies, but overall were documented for 30 to 36% of children treated with phenobarbitone or benzodiazepines.
What does current guidance say on this issue?
NICE guidelines on assessing and managing fever in under-fives (2013) emphasise that regular antipyretic drugs do not prevent febrile convulsions and should not be used for this purpose. When using paracetamol or ibuprofen in children who are distressed with a fever, NICE suggest continuing if the child appears distressed.
NICE’s Clinical Knowledge Summary gives guidance on the acute management of febrile convulsions, including advice on when a hospital assessment is required to exclude other causes of a convulsion. Drugs to manage or prevent further convulsions should not be prescribed unless advised by a specialist. Antipyretics are suggested, to reduce fever in future illness, but this may not prevent recurrence.
What are the implications?
This systematic review suggests that intermittent diazepam or continuous phenobarbitone may have some effect in preventing further febrile convulsions. However, this benefit was not large enough to outweigh the potential harms associated with these drugs. Other anticonvulsants are ineffective yet still carry potential harms.
Parents and carers may feel concerned about not receiving specific preventive treatment for these worrying, but predominantly harmless, convulsions. Therefore, there may be scope for creating materials for parents and carers about febrile convulsions, what they can do during a convulsion and why drug treatments are of little help.
Citation and Funding
Offringa M, Newton R, Cozijnsen MA, Nevitt SJ. Prophylactic drug management for febrile seizures in children. Cochrane Database Syst Rev. 2017;2:CD003031.
This review was funded by the Department of Pediatric Clinical Epidemiology, Emma Childrens’ Hospital A.M.C. Amsterdam, Netherlands and the Dutch Cochrane Centre, Amsterdam, Netherlands. Cochrane UK and the Epilepsy Cochrane Review Group are supported by NIHR infrastructure funding.
NHS Choices. Febrile seizures. London: Department of Health; 2015.
NICE. Febrile seizure. London: National Institute for Health and Care Excellence; 2013.
NICE. Fever in under 5s: assessment and initial management. CG160. London: National Institute for Health and Care Excellence; 2013.
Febrile convulsions occur in a substantial proportion of infants and they commonly re-occur. Studies have shown that prophylactic treatment can be effective in preventing the recurrence of convulsions. However, such prophylactic treatment can have unwanted effects in almost a third of the infants, and this is not widely known.
At the same time we know that, although scary when they happen, recurrent febrile convulsions are benign. We should therefore make more effort to explain to families the benign nature of these convulsions and provide information on how to act when a fever or febrile convulsion occurs.
Dr Brigitte Vollmer, Associate Professor of Neonatal and Paediatric Neurology, Honorary Consultant in Paediatric Neurology, University of Southampton