PEDIATRICS Vol. 122 No. 4 October 2008, pp. 731-735 (doi:10.1542/peds.2007-3278)
ARTICLE |
Prophylactic Phenobarbital Administration After Resolution of Neonatal Seizures: Survey of Current Practice
a Department of Pediatrics, Division of Neonatology
b Department of Neurology, Division of Child Neurology, Golisano Children's Hospital at Strong, University of Rochester Medical Center, Rochester, New York
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
OBJECTIVE. Child neurologists and neonatologists often discharge newborn infants with phenobarbital treatment for weeks to months despite the absence of continuing seizure activity. We conducted a national survey to determine the degree of variation in this practice.
METHODS. Surveys were sent to a randomly generated list of board-certified child neurologists (N = 609) and neonatologists (N = 579). The survey consisted of 3 parts, that is, questions related to overall attitudes and practices, specific patient scenarios, and respondent demographic characteristics. Responses were tabulated and analyzed for all respondents combined and for child neurologists and neonatologists separately. Variation in practices between respondents and the consistency between the respondents' stated use of phenobarbital in practice and their answers to various clinical scenarios were evaluated.
RESULTS. Responses were received from 118 child neurologists (20.7%) and 125 neonatologists (23.1%). There was wide variation in practices, with little difference in the response frequencies between child neurologists and neonatologists. Physicians were more likely to respond yes to continuation of phenobarbital treatment in a given clinical situation than would be predicted on the basis of their answers regarding overall frequency of use.
CONCLUSIONS. Since the survey of practices 15 years ago, child neurologists and neonatologists are reporting less frequent and shorter phenobarbital treatment after resolution of neonatal seizures, although there remains considerable variation in practices. Moreover, what physicians report as their practice in general is inconsistent with how they respond to specific clinical cases of neonatal seizures.
Key Words: neonatal seizures • phenobarbital • survey
Neonatal seizures are a common problem, affecting 1 to 4 per 1000 live births.1–3 Child neurologists and neonatologists generally agree on the appropriate evaluation and initial pharmacologic treatment of neonatal seizures. However, the best time to stop medication administration is not clear. Two surveys of practice, performed 10 years apart, illustrate the wide variation among specialties in discontinuing medications. Boer and Gal4 surveyed neonatologists and neurologists in the early 1980s and reported that the usual duration of treatment for patients with neonatal seizures varied from <1 month to 3 to 5 years. Relatively few providers (<15%) discontinued medications during the first month of life, and many (>35%) continued medications for >6 months. Small proportions (8% of neonatologists and 15% of neurologists) reported recommending 1 to 2 years of continued treatment or more. Similarly, in 1993, Massingale and Buttross5 reported that the duration of treatment ranged from 
3 months (40%) to 
1 year (11%) in a survey of neonatologists.
Informal discussions with neonatologists and child neurologists suggest that, although the frequency of continuation of phenobarbital treatment at the time of discharge may not be changing, the duration of treatment may be decreasing over time. Published surveys describing current practices are lacking. Therefore, we sought to determine the overall frequency with which term infants with neonatal seizures currently are discharged from the hospital with phenobarbital treatment, the usual duration of outpatient treatment for infants discharged from the hospital with phenobarbital therapy, and whether there are differences in practices as a function of the subspecialty training of the responsible physician and/or the presumed cause of the seizure. These data should help establish the study parameters for a future, randomized, clinical trial.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
A survey of practices was developed with the help of focus groups and experts in the area of neonatal neurology. Questions were pilot-tested for clarity and were revised after input from the reviewers. There were 3 distinct sections of the 9-page questionnaire, that is, 18 general questions regarding practices and attitudes, 14 questions on phenobarbital usage in specific clinical situations, and 15 demographic items.
A license was purchased from the American Board of Medical Specialties for use of a randomly generated list of board-certified child neurologists and neonatologists. Each physician on the list was sent a printed copy of the survey, with a cover letter offering them the option either to complete the printed copy of the survey or to complete the survey on-line by using the SurveyMonkey Web site (www.surveymonkey.com). A reminder postcard was sent to nonresponding physicians 
3 weeks after the initial mailing. The link to the SurveyMonkey questionnaire was included on the postcard. Three weeks later, a second printed copy of the survey was sent to everyone who had not yet completed the survey.
Survey data were compiled in a single spreadsheet by using SurveyMonkey and then were exported to Stata 9 (Stata, College Station, TX) for analysis. Responses were tabulated and analyzed for all respondents combined and for child neurologists and neonatologists separately. The responses to individual questions were evaluated by using standard summary statistics, and practice variations were quantified by using coefficients of variation. Correlations between several pairs of questions regarding the use of prophylaxis were analyzed by using McNemar's test. The ability of respondents to predict their own behavior in specific, hypothetical, clinical situations was evaluated by using receiver operating characteristic curve methods. These analyses were all conducted by using Stata. The University of Rochester institutional review board approved the study.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Of 770 child neurologists and 1696 neonatologists in the American Board of Medical Specialties database, 609 child neurologists and 579 neonatologists were randomly selected and sent surveys. Of those, 66 surveys were returned as undeliverable and 20 practitioners declined to participate. Responses were received from 118 child neurologists (20.7%) and 125 neonatologists (23.1%). Few respondents (9.5%) used the electronic option to complete the survey. Respondents represented a wide geographic area and work-setting spectrum (Table 1).
|
There was wide variation in practices, with little difference in the response frequencies for individual questions between child neurologists and neonatologists (Table 2). The majority of physicians who continue phenobarbital therapy after resolution of seizures prescribe it for
3 months, a smaller number suggest that phenobarbital therapy be continued for 3 to 6 months, and almost no one prescribes it for a longer period (Table 2). There was no difference overall or as a function of specialty in stated preferences to use prophylaxis as a function of geographic region, nursery size, presence or absence of pediatric residency, pediatric neurology, or neonatology training programs, number of infants cared for per month, or year of completion of subspecialty training. The coefficients of variation for the questions most relevant to clinical practice were >0.2, which supports the notion that there is wide variability in thinking and actual practices.
|
Furthermore, the responses to the question, "Do you typically discharge patients home on phenobarbital after resolved neonatal seizures?" did not predict how the respondents reacted to a given clinical scenario. Physicians were more likely to respond yes to continuation of phenobarbital treatment for a given scenario (Table 3) than would be predicted on the basis of their answers regarding the overall frequency of continuation of phenobarbital therapy (P < .01, McNemar's test). We used receiver operating characteristic methods to quantify further how well the responses to the questions on current practices predicted responses to questions related to a specific scenario. Surprisingly, the respondents' views of community practices seemed to be a slightly better predictor of how the respondents would react to specific scenarios. In no case was the area under the curve >0.75 (the value typically used to indicate a robust predictor6).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The decision to continue antiepileptic treatment after hospital discharge for infants with neonatal seizures is based on limited data. The rationale for using antiepileptic drugs at discharge is to decrease the likelihood of seizure recurrence. The rationale for not using prophylactic antiseizure treatment at discharge is either because of possible brain growth inhibition, as seen in animal models,7 or because of perceived low likelihood of seizure recurrence/timing of recurrence.3,8 The relevant data for neonates are limited to retrospective reviews, cohort studies of patients with neonatal seizures monitored prospectively for outcomes, with or without a standardized approach to outpatient care, and animal studies on the effects of chronic phenobarbital exposure on brain growth.7
Observational studies found no difference in seizure recurrence (epilepsy) when antiepileptic treatment was stopped early versus later.9,10 Overall, approximately one fourth of infants with neonatal seizures have 1 additional seizure in infancy or childhood, irrespective of the use of prophylaxis.11 Although risk assessment scores12 and electroencephalographic criteria13 have been proposed as determinants of the need for continued treatment, there is no evidence that either has been validated in other populations or widely adopted.
Farwell et al14 reported a randomized, clinical trial of prophylactic phenobarbital use in children 8 to 36 months of age with febrile seizures. Importantly, there was no difference at initial or later follow-up evaluations in the frequency of recurrent febrile seizures or later nonfebrile seizures.15 In addition, there were significant differences in subsequent neurodevelopmental assessments in the short term and in the longer term between the 2 groups, with the children who were treated with phenobarbital for 2 years experiencing worse outcomes. Because the prevention of seizure recurrence was the reason why children with febrile seizures were being treated with phenobarbital, this practice has been abandoned.
Compared with the results of surveys of practice performed 15 years5 and 25 years4 ago, the current survey confirms the trend toward shorter durations of phenobarbital use after resolution of neonatal seizures. However, this remains a widespread practice. Although practitioners may report that they never or rarely continue phenobarbital treatment after discharge, their inclination, when they are presented with specific clinical cases, may be to use phenobarbital as prophylaxis for recurrent seizures more often than not. The clinical vignettes in the survey were chosen to represent a range of severities and to mirror the spectrum and frequencies of causes seen in practice. Even allowing for the possibility that the scenarios included in the survey were more "serious" or complex than those seen in everyday practice, some physicians who answered never to the question on overall usage responded that they would continue phenobarbital treatment after discharge for up to 4 of the 7 children described. Although practice has evolved toward decreased duration of phenobarbital treatment after resolution of neonatal seizures, phenobarbital is still prescribed for a significant proportion of patients.
This survey was intended to reflect current practices of both neonatologists and pediatric neurologists in a wide variety of settings across the United States. We chose not to limit the participants to those in academic centers and thus did not target either training program directors or division chiefs. Instead, we chose to send the survey to a randomly selected group of board-certified physicians, in an attempt to decrease bias related to age, practice location or type, gender, or affiliation with training programs. Our response rate of 20%, despite multiple reminders, was lower than we hoped for. Although response rates generally are thought to be related to response bias, there is evidence that this may not be the case.16 We acknowledge that the survey was lengthy and demanded careful thought. This was brought up during pilot-testing of the survey, and several people at professional meetings admitted that this was a primary barrier to completion of the survey. Another comment heard repeatedly is that requests for surveys for specialists are now very commonplace and, over time, all forms of surveys have had lower response rates. Also, we chose not to compensate individuals for completion of the survey.
Examination of the characteristics of the respondents and nonrespondents in our study suggested that they were nearly identical in their geographic distribution. Other characteristics, such as whether the practitioners were in academic or private practice, were more difficult to compare, given the basic information (name and address) provided by the American Board of Medical Specialties. In our analyses, however, no aspect of work setting (academic affiliation, size of NICU, presence of training programs, or geographic location) or training (year of training) was associated with preferences to prescribe phenobarbital prophylaxis. Therefore, we think that the data from this sample of almost 250 physicians who currently care for infants with neonatal seizures are representative of overall (and admittedly inconsistent) practices at this time.
Animal models suggest that the neonatal brain is vulnerable to repeated seizures.17–20 However, exposure of the developing brain to phenobarbital for prolonged periods may have deleterious consequences.14,15,21 Extrapolation from the results of a randomized, clinical trial of phenobarbital prophylaxis after febrile seizures in children suggests that phenobarbital may be ineffective in preventing seizure recurrence and may affect brain development adversely.14,15 Therefore, a randomized, clinical trial is needed to determine whether outcomes are improved or impaired for infants who continue phenobarbital treatment in the absence of continuing seizure activity.
|
ACKNOWLEDGMENTS |
|---|
This survey was supported by a Clinical Trials Planning Grant (R34 HD050102) awarded to Dr Guillet.
We appreciate the time and effort of all those who provided input during development of the questionnaire and of Whitney Beck, a participant in the University of Rochester Summer Research Program for Undergraduates.
|
FOOTNOTES |
|---|
Accepted Jan 9, 2008.
Address correspondence to Ronnie Guillet, MD, PhD, Department of Pediatrics, Division of Neonatology, Box 651, Golisano Children's Hospital at Strong, University of Rochester Medical Center, 601 Elmwood Ave, Rochester, NY 14642. E-mail: ronnie_guillet{at}urmc.rochester.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
| What's Known on This Subject There is wide variation among practitioners regarding the continuation of phenobarbital treatment after resolution of neonatal seizures. Recurrent seizures may be deleterious to the developing brain, as may be prolonged exposure to phenobarbital.
|
| What This Study Adds This study provides data on the current use of phenobarbital after resolution of neonatal seizures among child neurologists and neonatologists. Considerable variation still exists, but the duration of treatment may be shorter than in the past.
|
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
- Rennie JM, Boylan GB. Neonatal seizures and their treatment. Curr Opin Neurol. 2003;16 (2):177 –181[CrossRef][Web of Science][Medline]
- Rennie JM. Neonatal seizures. Eur J Pediatr. 1997;156 (2):83 –87[CrossRef][Web of Science][Medline]
- Scher MS, Aso K, Beggarly ME, Hamid MY, Steppe DA, Painter MJ. Electrographic seizures in preterm and full-term neonates: clinical correlates, associated brain lesions, and risk for neurologic sequelae.
Pediatrics. 1993;91
(1):128
–134
[Abstract/Free Full Text] - Boer HR, Gal P. Neonatal seizures: a survey of current practice.
Clin Pediatr (Phila). 1982;21
(8):453
–457
[Abstract/Free Full Text] - Massingale TW, Buttross S. Survey of treatment practices for neonatal seizures. J Perinatol. 1993;13 (2):107 –110[Medline]
- Fan J, Upadhye S, Worster A. Understanding receiver operating characteristic (ROC) curves. CJEM. 2006;8 (1):19 –20[Medline]
- Diaz J, Schain RJ. Phenobarbital: effects of long-term administration on behavior and brain of artificially reared rats.
Science. 1978;199
(4324):90
–91
[Abstract/Free Full Text] - Holden KR, Mellits ED, Freeman JM. Neonatal seizures, part I: correlation of prenatal and perinatal events with outcomes.
Pediatrics. 1982;70
(2):165
–176
[Abstract/Free Full Text] - Gal P. Anticonvulsant therapy after neonatal seizures: how long should it be continued? Part I: a case for early discontinuation of anticonvulsants. Pharmacotherapy. 1985;5 (5):268 –273[Web of Science][Medline]
- Clancy RR, Legido A. Postnatal epilepsy after EEG-confirmed neonatal seizures. Epilepsia. 1991;32 (1):69 –76[Web of Science][Medline]
- Guillet R, Kwon J. Seizure recurrence and developmental disabilities after neonatal seizures: outcomes are unrelated to use of phenobarbital prophylaxis.
J Child Neurol. 2007;22
(4):389
–395
[Abstract/Free Full Text] - Ellison PH, Largent JA, Bahr JP. A scoring system to predict outcome following neonatal seizures. J Pediatr. 1981;99 (3):455 –459[CrossRef][Web of Science][Medline]
- Scarpa P, Chierici R, Tamisari L, Fortini C, Volpato S. Criteria for discontinuing neonatal seizure therapy: a long-term appraisal. Brain Dev. 1983;5 (6):541 –548[Web of Science][Medline]
- Farwell JR, Lee YJ, Hirtz DG, Sulzbacher SI, Ellenberg JH, Nelson KB. Phenobarbital for febrile seizures: effects on intelligence and on seizure recurrence. N Engl J Med. 1990;322 (6):364 –369[Abstract]
- Sulzbacher S, Farwell JR, Temkin N, Lu AS, Hirtz DG. Late cognitive effects of early treatment with phenobarbital.
Clin Pediatr (Phila). 1999;38
(7):387
–394
[Abstract/Free Full Text] - Cull WL, O'Connor KG, Sharp S, Tang SF. Response rates and response bias for 50 surveys of pediatricians. Health Serv Res. 2005;40 (1):213 –226[CrossRef][Web of Science][Medline]
- Haut SR, Veliskova J, Moshe SL. Susceptibility of immature and adult brains to seizure effects. Lancet Neurol. 2004;3 (10):608 –617[CrossRef][Web of Science][Medline]
- McCabe BK, Silveira DC, Cilio MR, et al. Reduced neurogenesis after neonatal seizures.
J Neurosci. 2001;21
(6):2094
–2103
[Abstract/Free Full Text] - Holmes GL, Ben-Ari Y. Seizures in the developing brain: perhaps not so benign after all. Neuron. 1998;21 (6):1231 –1234[CrossRef][Web of Science][Medline]
- Holmes GL, Gairsa JL, Chevassus-Au-Louis N, Ben-Ari Y. Consequences of neonatal seizures in the rat: morphological and behavioral effects. Ann Neurol. 1998;44 (6):845 –857[CrossRef][Web of Science][Medline]
- van der Pol MC, Hadders-Algra M, Huisjes HJ, Touwen BC. Antiepileptic medication in pregnancy: late effects on the children's central nervous system development. Am J Obstet Gynecol. 1991;164 (1):121 –128[Web of Science][Medline]
PEDIATRICS (ISSN 1098-4275). ©2008 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
