Fatemeh Ghotbi  ( Department of Pediatrics, Taleghani Medical Center, Shaheed Beheshti Medical University, Tehran, Iran. )
Farideh Shiva  ( Department of Pediatrics, Taleghani Medical Center, Shaheed Beheshti Medical University, Tehran, Iran. )

Objective: It is frequently thought that lumbar puncture (LP), is a mandatory procedure in all children who have fever and a seizure; because a convulsion may be the sole clinical manifestations of bacterial meningitis. To assess whether meningitis could be recognized using readily available clinical information.
Methods: This study was done during a 4 yr period from 2002-2006. A total of 254 previously healthy children aged 6 months to 5 years, were brought consecutively to the paediatric department of a teaching university hospital after their first fever-associated-seizure; lumbar puncture (LP) was performed in all cases. Children with seizure and fever and meningitis served as cases and those with fever and seizure, but no meningitis, served as control. Factors compared in the two groups were: age, lethargy, irritability, vomiting, nuchal rigidity, bulging fontanel, headache, drowsiness, toxicity, coma, complex seizure, and prior antibiotic use.
Results: Twelve, (4.7 %), cases were diagnosed as meningitis. Risk factors significantly associated with meningitis were: age <12 months, lethargy, irritability, vomiting, nuchal rigidity, bulging fontanel, headache, drowsiness, toxicity, coma, complex seizure, and prior antibiotic use, (p < 0.05). All children with meningitis had at least one of the risk factors mentioned above.
Conclusion: Our results indicate that based on available clinical data, meningitis can be ruled out in children presenting with seizure and fever; thus there is no need for routine lumbar puncture in all children who present with fever and seizure. However a lumbar puncture is mandatory in infants younger than 12 months or who have received prior antibiotics (JPMA 59:292; 2009).
Introduction
Seizure and fever are common findings among 2-5% children aged 4 months to 5 years   and has been reported in 25-30 % of children with bacterial meningitis.1-5 One of the most important clinical decision to make is whether a lumbar puncture, (LP) is necessary to rule out pyogenic meningitis or encephalitis.6,7 Several researchers are of the opinion that meningitis may be present in children presenting with seizure and fever in the absence of typical clinical manifestations; these investigators recommend a lumbar puncture in all children with seizure and fever except in those cases where the possibility of CNS infection seems truly remote.6-8 On the other hand, doing lumbar puncture (LP), has its hazards, i.e. in children with bacteraemia, it could end in meningitis.3,5 Other investigators, have questioned the necessity of doing LP in children with seizure and fever without signs of meningitis, and recommend that LP should only be done on children selected on the basis of history and clinical manifestations suggestive of meningitis.3,8,9 The incidence of bacterial meningitis as a cause of febrile seizures has decreased over the past 20 years and is now between 0.23% and 2%; recent retrospective studies have found that children with bacterial meningitis almost always demonstrate meningeal signs, focal neurologic findings, or signs of extreme illness.10 The risk of pyogenic meningitis in children with simple febrile seizures equals the risk of meningitis (<1.3%) in a febrile child without seizure.11 The object of this study was to see if we could differentiate children with meningitis, merely by relying on the history and physical examination alone.
Methods
This  study was done during a 4 year period from 2002-2006, on children aged 6 months to 5 years, who have been hospitalized in the paediatric ward at Taleghani Teaching Hospital associated with Shaheed Beheshti University of Medical Sciences. In these children, a complete history was taken upon hospitalization, and then, a physical examination was performed by the resident in charge, paying special attention to the signs of meningitis, i.e. lethargy, nuchal rigidity, bulging fontanel, drowsiness, or coma. The data was documented, and transferred to EXCEL software. Laboratory tests including CBC, electrolytes, urinalysis, urine culture, faecal tests and culture, and lumbar puncture were done. Also CSF tests for sugar, protein, WBC, and CSF culture were carried out in all patients. The child was considered as a case of meningitis if: WBC >10/cu mm, Gram stain positive for bacteria, and/or a positive CSF culture. Also, the CSF in cases with meningitis had a sugar level <40mg/dL and or protein >80mg/dL. Children with seizure and fever and meningitis, served as cases, and those with fever and seizure, but no meningitis were categorized as controls. Variables considered for comparison  in the two groups were: age, lethargy, irritability, vomiting, nuchal rigidity, bulging fontanel, headache, drowsiness, toxicity, coma, complex seizure, i.e. repeated seizure, seizure lasting for longer than 15 minutes, and prior antibiotic use. All data was analyzed using Fisher exact test, p value and odds ratio. Using Woolf approximation, a 95 % confidence interval was obtained.
Results
Two-hundred and fifty-four children between the ages of 6 to 60 months presenting with a first seizure with fever were included.  There were 137 boys (the ratio of boys to girls = 1.2: 1). Average age was 19.3 months. In 12 cases, meningitis was diagnosed, 5 were bacterial and the rest were viral meningitis or encephalitis. Among those who had meningitis seven had received antibiotic as outpatients prior to hospitalization, in two of them the only clinical finding was fever. Table 1 shows the characteristics of the children with and without meningitis. Fifty-nine cases were under 1 year of age. There was a significant difference in the incidence of meningitis in children under and over 12 months of age, i.e. 11% vs. 2.56%, respectively, (p= 0.0078). Three cases, (25 %), of those with, and 27 cases, (9.9 %), without meningitis, had seizure lasting longer than 15 minutes, which statistically was not meaningful. Six cases, (50 %), with meningitis, and 53 cases, (21 %), without meningitis, had more than one seizure, which was statistically significant, (Table 2).  Seven cases, (58 %), with, and, 27 cases, (11 %), without meningitis, had received antibiotics as outpatients prior to admission, this was statistically meaningful, (Table 2). Table 2 shows clinical findings and their statistical analysis of the children with and without meningitis. The etiology of fever in children in our study is as follows: URI in 110 cases, gastroenteritis in 73 cases, URI and otitis media in 35 cases, meningitis in 12 cases, five of which were bacterial. UTI in 10 cases, gastroenteritis and URI in 7 cases, roseola in 4 cases,  infectious mononucleosis  in 2 cases, and streptococcal pharyngitis in 1 case.
Discussion
Meningitis was diagnosed in 4.7 % of the children in our study. These numbers compare well with previous studies performed by Akpede5 in which  4.2% of 522 children between ages of 1-6 months presenting with fever and seizure were diagnosed as meningitis and Al-Essai12 who reported 3.5% of their patients as having meningitis; however Carrol et al3 are of the opinion that  the risk of bacterial meningitis is extremely small (less than one in two hundred), even in children under 1 year of age who are hospitalized with a febrile convulsion. In contrast Offringa's study on 309 children with fever and seizure reports a higher incidence, (7%), of meningitis in their patients.9 These figures are in stark contrast to those quoted by Teach and colleagues who claim that in a study done on 243 children admitted due to seizure and fever, without any clinical signs of meningitis, none of them had meningitis.13 Other studies that show the incidence of bacterial meningitis in children who presented to the emergency room because of their first febrile seizure but with no clinical signs of meningitis are as follows: Dawson and Capaldi,14 (100 patients, 0 meningitis); Anderson et al,15 (100 patients, 0 meningitis); Von Stuijvenberg,16 (203 cases, 0 meningitis); Kinsella et al,17 (47 patients, 0 meningitis). However  there have been reports of missing meningitis in children with febrile seizures, notably a case report of a 12-month old infant with acute bacterial meningitis with a simple febrile seizure with no other signs.10 According to other reports, the frequency of bacterial meningitis in a child with seizures and fever is low, but nevertheless significant, with studies showing up to 7% of children having bacterial meningitis; most children with meningitis who present with fever and seizures have an altered conscious state, signs of increased intracranial pressure, or signs of meningism, but it should be remembered that meningeal signs can be absent in infants and young children with meningitis.18 In our patients the following risk factors were associated with meningitis: age under 1 year, repeated seizures, onset of seizure 24 hours after the onset of fever, seizures lasting longer than 15 minutes, drowsiness, nuchal rigidity, irritability of unknown origin, lethargy, vomiting, bulging fontanel, headache, and finally, the use of antibiotics prior the onset of seizure. A previous survey reported the discriminating signs between children with and without meningitis as: petechiae, nuchal rigidity, coma, continuous drowsiness, ongoing seizure in the emergency room, paresis, and paralysis. Children whose seizures showed no complex features and whose febrile illness revealed no suspicious features did not have meningitis. The results of these investigators showed that based on clinical data, meningitis can be ruled out in children presenting with seizure and fever so there is no need for a routine LP8 In Akpede's5 study six children who had bacterial meningitis, had none of the classic signs of meningitis, but had other indications for an LP, including: age < 6 months, focal or repeated seizures, absence of seizure in the familial history, coma, and presence of a source of infection outside the CNS. These researchers believed the decision to do an LP, will depend on such factors as age, and complex seizures with fever. In Lorber's19 study clinical signs observed to be predictive of meningitis were: prolonged illness, photophobia, nuchal rigidity, Kerning's sign, continuation of fever without any specific cause, and deterioration of patient´s condition in hospital.  In two of our patients classical signs of meningitis were absent; however one of these patients was only eight months of age and the other, who was 23 months old, had experienced prolonged seizure lasting >15 minutes and both  these infants had received oral antibiotics prior to the onset of seizure. Our observations were similar to those found in Al-Essai’s12 study, where factors as age less than 18 months, fever lasting more than 48 hours, questionable signs of meningitis, and complex seizures, were quoted as being important in deciding to do an LP. According to Carrol and Brookfield,3 routine lumbar puncture procedure following a febrile convulsion in infancy is unjustified and potentially hazardous. Taking into consideration the above argument their practice following a febrile convulsion in infancy follows a "less is more" rule. Infants are assessed by a middle grade paediatrician shortly after admission and those without meningeal signs (irritability, lethargy, or bulging fontanels) who have recovered from their seizure are reviewed after four hours. If no deterioration has occurred and the child appears well, lumbar puncture is considered unnecessary.3 Offringa and colleagues20 who investigated 344 children with first experience of fever and seizure, in a prospective study concluded that doing LP, on patients with major clinical signs of meningitis, will mean that, only 16 children out of 304 (5%), would undergo a lumbar puncture, and, no cases of meningitis will be ignored. Moreover, only 2% of children will undergo an unnecessary lumbar puncture. Similar to some other studies, almost all our patients with meningitis had signs of meningeal irritation viz. irritability or/and nuchal rigidity.3,8,18,19 Another finding that was noticed in our study was the development of complex febrile seizures in most of the children with meningitis, a risk factor that has been quoted by some other authors as well.5,8,12 Approximately one-third of our patients presented with altered consciousness, a finding that has been reported previously.8,18 An additional risk factor in our patients was prior use of antibiotics; about 58% of meningitis cases in our study had used oral antibiotics before hospitalization as compared to 11% without meningitis. Our results are consistent with the reports from researchers in the field who are of the opinion that the risk of meningitis is negligible in children with a simple febrile seizure in the absence of risk factors such as age younger than 18 months, prior use of antibiotics or sings of meningeal irritation.3-8,12-17
Conclusion
Our findings suggested that in children presenting with seizure and fever, spinal tap should be reserved for cases with clinical manifestations of meningitis, or the presence of risk factors such as age <12 months, repeated seizures and/or prior use of antibiotics. In addition, a delayed convulsion i.e. more than 24 hours after the onset of fever constitutes an independent risk factor for meningitis. 
References
1.Teng D, Dayan P, Tyler S, Hauser WA, Chan S, Leary L, et al. Risk of intracranial pathologic conditions requiring emergency intervention after a first complex febrile seizure episode among children Pediatrics 2006; 117:528-30.
2.Duffner PK, Baumann RJ, Berman P, Green JL, Schneider S, Hodgson ES, et al. Febrile seizures: clinical practice guideline for the long-term management of the child with simple febrile seizures. Steering Committee on Quality Improvement and Management, Subcommittee on Febrile Seizures American Academy of Pediatrics. Pediatrics 2008; 121:1281-6.
3.Carroll, D Brookfield.Lumbar Puncture Following Febrile Convulsion. Arch Dis Child 2002; 87: 238-40.
4.Fetveit A. Assessment of febrile seizure in children. Eur J Pediatr 2008; 167:17-27.
5.Akpede GO, Sykes RM, Abiodun PO. Indication for lumbar puncture in children presenting with convulsions and fever of acute onset: experience in children´s emergency room of the University of Benin Teaching Hospital, Nigeria. Ann Tropical Pediatrics 1992; 12:385-9.
6.Practice parameter. The neurodiagnostic  evaluation of the child. American Academy of Pediatrics. Provisional Committee on Q1. Febrile Seizures Pediatrics 1996; 97:769-72.
7.Baumann RJ, D´Angelo SL, Technical report summary: The neurodiagnostic evaluation of the child with a first simple febrile seizure. Pediatrics 1996; 97:773-5.
8.Rosman NP. Evaluation of the child who convulses with fever. Paediatr Drugs 2003; 5:457-61.
9.Offringa M, Beishuzen A, Lubsen GD, Lubsen J. Seizures and fever: can we rule out meningitis on clinical grounds alone? Clin Pediatr (Phila) 1992; 31:514-22.
10.Golnik A. Pneumococcal meningitis presenting with a simple febrile seizure and negative blood-culture result. Pediatrics 2007; 120:e428-31.
11.Karande S. Febrile seizure: a review for family physician. Indian J Medicine 2007; 61: 161-72.
12.AL-Eissa YA. Lumbar puncture in the clinical evaluation of children with seizures associated with fever. Pediatr Emerg Care 1995; 11: 347-50.
13.Teach SJ, Geil PA. Incidence of bacteremia, urinary tract infections, and unsuspected bacterial meningitis in children with febrile seizure. Pediatr Emerg Care 1999; 15:9-12.
14.Dawson KP, Capaldi NI.  The initial hospital management of childhood febrile convulsions: an audit.J Qual Clin Pract 1994; 14:111-4.
15.Anderson AB, Desisto MJ, Marshal PC, Dewitt TG. Duration of fever during onset of a simple febrile seizure: a predictor of significant illness and neurologic course. Pediatr Emerg Care 1989; 5:12-5.
16.Van Stuijvenberg M, Moll HA, Steyerberg EW, Van Gijssel EN, Moons KG, Derksen-Lubsen G. The duration of febrile seizure and peripheral leukocytosis. J Pediatr 1998; 133:557-8.
17.Kinsella JB, O´sullivan P, Mcshane DP. The role of the middle ear and tonsil in the etiology of febrile convulsion. Int J Pediatr Otorhinolaryngol 1995; 32:153-7. 18.Srinivasan J, Wallace KA, Scheffer IE. Febrile seizures. Australian Family Physician 2005; 34:1021-5.
19.Lorber J, Sunderland R. Lumbar puncture in children with convulsions associated with fever. Lancet 1980; 1:785-6.
20.Offringa M. Seizures associated with fever, current management controversies.  Semin Pediatr Neurol 1994; 1:90-101.