Which children with febrile seizures need lumbar puncture? A decision analysis approach

Whether all children brought to the emergency room with a first seizure and fever require lumbar puncture (LP) remains controversial. We reviewed the emergency room records of 241 children aged 6 months to 6 years who had this clinical picture. Five history and physical examination items discriminated between children with and without meningitis: a physician visit within 48 hours before the seizure; the occurrence of convulsions on arrival at the emergency room; a focal seizure; suspicious findings on physical and/or neurologic examination. Used in combination, these items (risk factors) identified all children with meningitis but would have spared 62% of children without meningitis the need for LP. In a decision analysis framework, they were as sensitive but more specific than LP in detecting children with meningitis. Most important, their negative predictive value was 100%.     

Diseases that mimic meningitis. Analysis of 650 lumbar punctures

A retrospective review of charts for 650 children who had lumbar puncture for suspected meningitis was undertaken to determine the characteristics of patients with and without meningitis, identify other conditions suggesting meningitis, and evaluate the predictive value of signs and symptoms of meningitis. The incidence of positive lumbar punctures increased with patient age. Younger infants did not present with classical features of meningitis. Bulging fontanel, lethargy, and irritability were nonspecific symptoms. Vomiting and headache, although not specific, proved to be more sensitive indicators of meningeal infection. Most patients with meningitis (75%) had at least one sign of meningeal irritation, but so did 25% of patients without meningitis. Brudzinski's sign was not specific. In contrast, nuchal rigidity and Kernig's sign had high predictive value. Up to age five, the diseases most often suggesting meningitis were right-sided pneumonia, gastroenteritis, otitis, tonsillitis, exanthema subitum, and urinary tract infections. Of 171 patients with febrile convulsion, one (0.5%) had bacterial meningitis and four had aseptic meningitis.     

Role of routine investigations in children presenting with their first febrile convulsion.

To assess the role of routine investigations in children presenting with their first febrile convulsion, the results of investigations carried out in 328 children over a 2-year period were reviewed. Lumber puncture was performed in 96% of cases and resulted in the detection of 4 cases of unsuspected meningitis, one of which was bacterial. 2 children had normal lumbar punctures on admission but developed meningococcal meningitis within 48 hours. Sugar, calcium, urea, and electrolyte estimations, and blood counts were commonly performed but were unhelpful. We suggest that lumbar puncture in those children presenting with their first febrile convulsion under the age of 18 months is the only useful routine investigation.     

Role of routine investigations in children presenting with their first febrile convulsion

To assess the role of routine investigations in children presenting with their first febrile convulsion, the results of investigations carried out in 328 children over a 2-year period were reviewed. Lumber puncture was performed in 96% of cases and resulted in the detection of 4 cases of unsuspected meningitis, one of which was bacterial. 2 children had normal lumbar punctures on admission but developed meningococcal meningitis within 48 hours. Sugar, calcium, urea, and electrolyte estimations, and blood counts were commonly performed but were unhelpful. We suggest that lumbar puncture in those children presenting with their first febrile convulsion under the age of 18 months is the only useful routine investigation.     

Lumbar puncture and febrile convulsions

We report the data of 878 selected children between 1 month and 6 years, presenting a first episode of seizure with fever. Two-hundred-fifty-five children underwent lumbar puncture. In 7 cases the CSF findings showed a bacterial meningitis, in 14 cases a viral meningitis. In 598 of the 623 children who did not undergo LP, a bacterial meningitis could be excluded on the basis of the clinical course. The data show that the probability of finding a bacterial or viral meningitis is high in children under 6 months of age even if no significant neurological signs are found on examination performed shortly after the seizure. In our study, older children affected by bacterial meningitis were clinically identifiable. In children aged 6 months to 3 years without important neurological signs, a complex seizure has been found to be a significant discriminating factor between patients with and without viral meningitis.     

Colloid cyst of the 3rd ventricle as a cause of acute neurological deterioration and sudden death

Abstract  A 13 year old girl presented with a 24 h history of headache and vomiting. There were no focal neurological signs. The diagnoses considered were vascular headache, meningitis and subarachnoid haemorrhage. Lumbar puncture revealed clear cerebrospinal fluid under high pressure and subsequent cerebral computed tomography (CT) scan revealed hydrocephalus. Death occurred some hours after the lumbar puncture. Autopsy revealed a colloid cyst of the 3rd ventricle causing the hydrocephalus. Cerebellar tonsillar herniation was present.
This case is reported because although 3rd ventricular colloid cysts are a recognized cause of acute neurological deterioration and sudden death, they rarely present in childhood. In this case, it is considered that lumbar puncture may have hastened death by increasing brain-stem compression due to cerebellar tonsillar herniation. It raises the question of whether cerebral CT scan, where readily available, should be performed prior to lumbar puncture. Pathologically, colloid cysts may be easily missed if the brain is examined fresh as the cysts are fragile and have a tendency to be destroyed or fall out.     

Prediction of bacterial meningitis in children with meningeal signs: reduction of lumbar punctures

Physicians often have to perform a lumbar puncture to ascertain the diagnosis in patients with meningeal signs, because of the serious consequences of missing bacterial meningitis The aim of this study was to derive and validate a clinical rule to predict bacterial meningitis in children with meningeal signs, to guide decisions on the performance of lumbar punctures. Information was collected from records of patients (aged 1 mo to 15 y) consulting the emergency department of the Sophia Children's Hospital between 1988 and 1998 with meningeal signs. Bacterial meningitis was defined as cerebrospinal fluid (CSF) leucocyte count >5 cells μl^?1 with a positive bacterial culture of CSF or blood. The diagnostic value of predictors was judged using multivariate logistic modelling and area under the receiver operating characteristic curves (ROC area). In the derivation set (286 patients, years 1988–1995) the duration of the main complaint, vomiting, meningeal irritation, cyanosis, petechiae and disturbed consciousness were independent clinical predictors of bacterial meningitis. The ROC area of this model was 0.92. The only independent predictor from subsequent laboratory tests was the serum C‐reactive protein concentration, increasing the ROC area to 0.95. Without missing a single case, this final model identified 99 patients (35%) without bacterial meningitis. Validation on 74 consecutive patients in 3 subsequent years (1996–1998) yielded similar results. Conclusion: This prediction rule identifies about 35% of the patients with meningeal signs in whom a lumbar puncture can be withheld without missing a single case of bacterial meningitis. For the individual patient this prediction rule is valuable in deciding whether or not to perform a lumbar puncture.     

Assessment of febrile seizures in children

Febrile seizures are the most common form of childhood seizures, affecting 2–5% of all children and usually appearing between 3 months and 5 years of age. Despite its predominantly benign nature, a febrile seizure (FS) is a terrifying experience for most parents. The condition is perhaps one of the most prevalent causes of admittance to pediatric emergency wards worldwide. FS, defined as either simple or complex, may be provoked by any febrile bacterial or (more usually) viral illness. No specific level of fever is required to diagnose FS. It is essential to exclude underlying meningitis in all children with FS, either clinically or, if any doubt remains, by lumbar puncture. There is no evidence, however, to support routine lumbar puncture in all children admitted with simple FS, especially when typical clinical signs of meningitis are lacking. The risk of epilepsy following FS is 1–6%. The association, however small, between FS and epilepsy may demonstrate a genetic link between FS and epilepsy rather than a cause and effect relationship. The effectiveness of prophylactic treatment with medication remains controversial. There is no evidence of the effectiveness of antipyretics in preventing future FS. Prophylactic use of paracetamol, ibuprofen or a combination of both in FS, is thus a questionable practice. There is reason to believe that children who have experienced a simple FS are over-investigated and over-treated. This review aims to provide physicians with adequate knowledge to make rational assessments of children with febrile seizures.     

Septicemic neonates without lumbar puncture: what are we missing?

Meningitis is a serious problem in newborn infants and has high mortality and frequent neurological sequelae. In neonates, signs and symptoms of serious infections are often obscure and clinical examination cannot distinguish septicemic babies with or without meningitis. Therefore, lumbar puncture is often not done in time and thus diagnosis of meningitis is missed. This study aimed to discover the prevalence of meningitis among these cases based on laboratory investigation. We prospectively enrolled the blood culture positive septicaemia cases which were not labeled as cases of meningitis during routine clinical evaluation. Out of 30 septicemic cases, eight (26.7%) had abnormal CSF cytology and biochemistry suggestive of meningitis. Among these eight cases, four had positive CSF culture; [Klebsiella pneumoniae (n = 3) and Pseudomonas aeruginosa (n = 1)], which were similar to the blood isolate of the respective patient. The clinical manifestations were similar in both septicemia and meningitis cases. Mortality was high among the meningitis cases compared with those having septicemia alone (37.5% vs. 13.3%), indicating the need for early diagnosis of this disease. Our data confirmed that it is important to do a lumbar puncture, along with blood culture, for all suspected septicemia cases.     

Neisseria meningitidis bacteremia in children: quantitation of bacteremia and spontaneous clinical recovery without antibiotic therapy

The relationship between the magnitude of bacteremia due to Neisseria meningitidis and the clinical diagnosis was determined for 43 children who had fever in the presence or absence of focal signs of infection. Bacteremia was quantitated by the previously described procedure using heparinized blood (0.2 to 1.0 mL). Additionally, blood was cultured by means of the radiometric Bactec technique. Seventeen patients had meningitis, 12 had meningococcemia, 13 had unsuspected or "occult" bacteremia, and five had other diagnoses. "Occult" bacteremia was diagnosed initially in four patients, but subsequently meningitis was diagnosed. All 13 patients with 500 or more organisms per milliliter had meningitis or meningococcemia in contrast to 12 (55%) of 22 patients with less than 500 organisms per milliliter (P less than or equal to .0035). Only 18 (42%) of these patients bacteremic with N meningitidis presented with petechiae or purpura. All 13 children with occult bacteremia were sent home after blood cultures were obtained; six of the 13 received a regimen of oral amoxicillin for otitis media. At reexamination (interval 16 to 119 hours) four had meningitis, seven were clinically improved (afebrile, negative blood culture, without invasive disease), and two were still mildly febrile with negative blood culture. Three of these bacteremic children experienced spontaneous clinical and bacteriologic resolution without antibiotic treatment. This has not been previously reported.     

Clinical significance of enteroviruses in serious summer febrile illnesses of children.

BACKGROUND: Enteroviruses are common causes of aseptic meningitis and nonspecific febrile illnesses in young children. During the summer-fall months, enterovirus-infected children are frequently evaluated in emergency room settings to rule out bacterial sepsis and/or meningitis. OBJECTIVES: We sought to determine the clinical significance of enterovirus infections in children evaluated for serious febrile illnesses in pediatric emergency rooms during the summer-fall season. METHODS: Children admitted to emergency rooms at four university teaching hospitals during a single summer-fall season who required blood culture and/or lumbar puncture to rule out bacterial sepsis/meningitis were prospectively studied. An extensive questionnaire was administered, and specimens of cerebrospinal fluid, serum, urine and throat were tested for enteroviruses by viral culture and PCR. Patients were followed to determine the duration, management and outcome of their illnesses. RESULTS: Of 203 patients studied 173 had no apparent explanation for their illness (e.g. bacterial sepsis, bacterial urinary tract infection, etc.). Of those 173 patients 79 (46%) were infected with enteroviruses, including 33 of 47 (70%) patients with aseptic meningitis, 13 of 25 (52%) patients with nonspecific febrile episodes and 33 of 101 (33%) patients with fever and focal findings (P < 0.0001 for aseptic meningitis vs. fever and focal findings; P = 0.0001 for aseptic meningitis vs. combined nonspecific febrile episodes and fever/focal patients). Among 119 hospitalized patients 65 (55%) were enterovirus-infected. Children < or =90 days of age were more likely to be enterovirus-infected (66 of 122; 54%) than children older than 90 days (13 of 51; 25%) (P = 0.0001). Enterovirus-infected children were more likely to be hospitalized as a result of the current emergency room visit (65 of 79 vs. 54 of 94; P = 0.0005) and were more likely to have had an additional hospitalization for the same illness (10 of 79 vs. 1 of 94; P = 0.003). Enterovirus-infected patients also had a shorter period from illness onset to presentation. Enterovirus-infected children were indistinguishable from those without enterovirus infection in their symptoms at onset, signs at presentation and total duration of illness (>7 days in both groups). Enterovirus-infected children were almost all treated with antibiotics (78 of 79; 99%), with 74 of 79 (94%) receiving parenteral antibiotics for a mean of 3.6 days. CONCLUSIONS: During the summer-fall months, 39% (79 of 203) of children for whom blood cultures and/or lumbar punctures were performed for suspected bacterial infection had enterovirus infection identified as the only explanation for their illness. Of those patients with no alternative diagnosis, enterovirus infection was confirmed in 46% (79 of 179). The majority of those patients requiring hospitalization were infected with enteroviruses. The use of PCR increases the number of children for whom a specific etiology of illness can be determined and may in the future reduce the hospitalization and use of unnecessary antibiotics in patients with enterovirus infections.     

Cerebral infarction in Hemophilus influenzae type B meningitis

Over a 5-year period, 8 (4.7%) of the 170 children diagnosed at Milwaukee Children's Hospital as having Hemophilus influenzae type b (HITB) meningitis developed cerebral infarction. Compared with children who did not develop infarcts or with children who developed other neurologic complications, such as subdural effusion, empyema, or meningoencephalitis, these children had significantly higher cerebrospinal fluid (CSF) leukocyte counts on initial lumbar puncture and had a greater likelihood of seizure activity. In seven of eight patients with cerebral infarction, a focal or generalized seizure heralded neurologic findings associated with abnormal radiographic studies. Two of the eight patients died, and two were permanently severely damaged. In the other four patients, there was eventual recovery from gross neurologic deficits. The mortality in patients with HITB meningitis complicated by cerebral infarction (25%) was significantly greater than that in other patients with HITB meningitis (0.6%). The pathophysiology of infarction in patients with bacterial meningitis is uncertain but may in part relate to arteriospasm. Cerebral infarction is a serious, and in the present experience, not uncommon complication of H. influenzae meningitis.     

Clinical indicators for lumbar puncture

This study was conducted to demonstrate that experienced pediatricians using standard clinical indications for performing a lumbar puncture should have a higher yield of positive spinal taps than previously reported and also can detect bacterial meningitis. These indicators included temperature elevation, inability to be consoled, level of alertness, nuchal rigidity, bulging fontanel, decreased appetite, rash, referral, and febrile seizures. Eighty-two of 381 (22%) lumbar punctures were positive for pleocytosis and/or organisms. Patients were divided into two groups, consisting of those with one indicator (low risk) and those with greater than one indicator (high risk). Thirteen of 14 patients with bacterial meningitis were placed in the high risk group. The single patient in the low risk group had been pretreated with antibiotics. The positive predictive value in bacterial meningitis for a score greater than one was 5%. The average number of clinical indicators in bacterial meningitis was 3.7, versus 2.4 in viral meningitis and 1.6 without meningitis. These findings suggest that, in the absence of prior antibiotic therapy, an experienced pediatrician can clinically detect patients at high risk for bacterial meningitis. Nonbacterial meningitis cannot be as readily detected clinically.     

Peritonitis as a first manifestation of Neisseria type C meningitis]

A 15-year-old boy was first referred for a clinical presentation of revealed peritonitis. Abdominal endoscopy showed normal appendix but the presence of purulent peritoneal fluid. Antibiotics were immediately administered. Ten hours later clinical signs of meningitis occurred, and lumbar puncture was performed. CSF bacterial meningitis characteristics were present but no bacteria was observed. However Neisseria meningitidis type C was detected by direct peritoneal fluid examination and by 24(th) hour blood culture. Peritoneal and the CSF fluid culture were negative, but DNA analysis from peritoneal fluid was positive.     

Meningitis following lumbar puncture in bacteremia?

We report on two infants with bacteremia and the clinical signs of meningitis who developed overt meningitis following an extended period after the first lumbar puncture, in which inconspicuous (cytological and bacteriological) cerebrospinal fluid were found. The causative infective agents of bacteremia and meningitis were identical. The iatrogenic induction of meningitis by the procedure of lumbar puncture in the presence of a bacteremia is imaginable, but has not been proven. Diagnostic consequences are: Every lumbar puncture for bacterial meningitis has to be complemented with a simultaneous blood culture. Lumbar puncture should be repeated when clinical signs of meningitis persists in children, especially in infants with positive blood culture and with inconspicuous cerebrospinal fluid findings in the initial lumbar puncture. Such children should be hospitalized for clinical observation. Therapeutical consequences are presently unclear. Proven recommendations are lacking for the different possible procedures after initial lumbar puncture in children with suspicion of a bacteremia: for instance no therapy or one single antibiotic dose given i.v. immediately following the lumbar puncture or the same treatment as in proven meningitis until microbiological results are obtained.     

Association of meningitis with infantile gastro-enteritis

It is the policy at the Jordan University Hospital to perform lumbar puncture on children with gastroenteritis who present with one or more of the following: age less than 1 month, convulsions, hypoactivity or marked irritability, and depressed sensorium. Review of the records of 737 children admitted with gastro-enteritis between January 1980 and October 1984 showed that lumbar puncture was performed on 351 (47.6%) children. Acute bacterial meningitis was diagnosed in only three children, two of whom had already received treatment before admission and the third had obvious meningeal signs. These findings do not justify the present policy on lumbar puncture in children with gastroenteritis and it is proposed that the procedure be reserved for children in whom abnormal CNS findings persist after initial correction of fluid and electrolyte balance or with overt signs of meningitis.     

Postictal cerebrospinal fluid abnormalities in children

OBJECTIVES: To determine the frequency and characteristics of seizure-induced cerebrospinal fluid (CSF) abnormalities in children and to identify potential alternative causes of these findings. METHODS: Consecutive patients (n = 80) who underwent lumbar puncture within 24 hours after a seizure were studied retrospectively. The presence of CSF abnormalities in total leukocytes, polymorphonuclear cells, and protein was determined by using age-specific reference values. Coexisting conditions that could affect CSF findings, such as traumatic lumbar puncture, concurrent neurologic disease, and undiagnosed meningitis, were identified. RESULTS: Eighteen of the 80 patients were excluded from the final study group because of the presence of another condition that could alter the CSF. More than 50% of the excluded patients had an abnormal CSF leukocyte count or protein level, including 2 patients with initially undiagnosed meningitis, which was subsequently detected by post-hoc polymerase chain reaction testing. In the remaining 62 patients, postictal pleocytosis was detected in only 3 (5%), and increased protein was detected in only 6 (10%). The maximal postictal pleocytosis and protein level were 8 x 10(6) leukocytes/L (8 leukocytes/mm(3)) and 0.52 g/L (52 mg/dL), respectively. CONCLUSIONS: Seizure-induced CSF abnormalities are rare in children, and alternative, often unidentified, disease processes may account for many observed postictal abnormalities. All patients with abnormal CSF after a seizure should be thoroughly evaluated for other causes of the abnormality.     

Duration of fever prior to onset of a simple febrile seizure: a predictor of significant illness and neurologic course

This study tested the hypothesis that the duration of fever prior to the onset of a simple febrile seizure may be an important clinical variable with respect to patient outcome. The duration of fever prior to seizure according to patient history was defined as either long (greater than or equal to 24 hours) or short (less than 24 hours). We hypothesized that simple febrile seizures which occur with a history of a fever of long duration (LDF) are more likely to be associated with a significant illness at presentation or a subsequent neurologically abnormal course than are simple febrile seizures which occur with a history of a fever or short duration (SDF). Of 100 cases which met study criteria for simple febrile seizures, nine had a LDF and 91 had a SDF prior to the development of a seizure. No statistical differences in age, sex, maximum fever recorded in the emergency department, duration of seizure, WBC, or electrolytes were found between patients with SDF and LDF (P less than 0.01). Of the nine patients with a LDF, all had either a significant illness at the time of initial visit or a subsequent neurologically abnormal course. Of the 91 patients with a SDF, 88 had a good outcome, while two had a significant illness at the time of visit, and one had a subsequent neurologically abnormal course. These results suggest that children with a history of LDF prior to the occurrence of a simple febrile seizure are more likely to have a serious illness at presentation or a subsequent neurologically abnormal course than are children with seizures which occur with a history of SDF.     

Cerebral cysticercosis.

A case of cerebral cysticercosis in an American child is described. The patient was only 2 years old and had never left the United States. Her symptoms began with febrile seizures and progressed to focal motor seizures. Cerebrospinal fluid pleocytosis with eosinophilia, candle-guttering of the walls of the ventricles on pneumoencephalography, and a titer of 1:4,096 against cysticercosis antigen in her blood led to the diagnosis. Over a five-year follow-up period, the patient's course has been one of resolution of her symptoms, improvement in her electroencephalogram, and excellent seizure control with anticonvulsant therapy. Cysticercosis should be considered in the differential diagnosis of a child who shows CSF pleocytosis with eosinophilia, particularly if accompanied by focal seizures.