a b s t r a c t

Background: In infants aged 3-18 months presenting with a bulging fontanelle and fever it is often necessary to exclude central nervous system infection by performing a lumbar puncture. Several studies have shown that well-appearing infants with normal clinical, laboratory and imaging studies have a benign (non-bacterial) dis- ease. At our institution, we often observe such infants and withhold lumbar puncture.

Objective: To determine the clinical characteristics and outcomes of well-appearing, febrile infants with a bulging fontanelle, whether they did or did not undergo lumbar puncture.

Design: A retrospective chart review of the medical records of all febrile infants with a bulging fontanelle seen be- tween March 2018 and March 2020 at Dana Children’s Hospital. The following data were extracted: age of the patient, gender, previous medical history, general appearance, vomiting, appetite, fever, blood test results and CSF results (when taken), final diagnosis, disposition status, and whether or not the patient returned to our ER. Stats: Descriptive statistics were used to describe the study population.

Results: The study group included 40 children, 22 males, and 18 females. Their age ranged from 3 to 13 months. Only 8 of the patients in the study group were admitted and 32 were discharged. Only 13 (32.5%) had an LP per- formed, three of which had elevated levels of WBC in the CSF, two of them had a positive culture. None of the pa- tients who were discharged returned to the ED.

Conclusions: Our study, combined with previous works, supports the assumption that management of well- appearing infants with normal medical history who present with fever and bulging fontanelle could be done without a lumbar puncture. Larger and prospective studies are needed to support this observation.

(C) 2022 Published by Elsevier Inc.

1. Background

Bacterial meningitis remains a life-threatening infection in children, even though its incidence has declined since the introduction of conju- gate vaccines against H. influenza And pneumococci [1,2]. Clinical signs are non-specific, making clinical Diagnosis difficult. Even though several studies have shown that well-appearing infants with normal clinical, laboratory and imaging studies, and a bulging fontanelle, have benign (non-bacterial) disease [3,4], some physicians continue to consider lumbar puncture mandatory in any febrile infant with a bulging fontanelle [5,6]. Several conditions can be associated with fever and a

* Corresponding author at: Division of Pediatric Emergency Medicine, Dana-Dwek Children Hospital, Sackler School of Medicine, Tel Aviv University, 6 Weizman Street, Tel- Aviv 64239, Israel.

E-mail address: [email protected] (M. Glatstein).

¹ The first two authors contributed equally to this study.

transient bulging fontanelle, including upper respiratory tract infection, roseola infantum, and Otitis media [7]. It is thought that these non- specific infective illnesses may interfere with absorption of the cerebro- spinal fluid by the arachnoid villi, resulting in transient Intracranial hypertension [8].

In our emergency department (ED) the decision to perform a LP is left up to the individual physician, prompting us to perform this retro- spective study examining outcomes of febrile infants with bulging fon- tanelle. We hoped that our experience would help inform the practice of physicians caring for such patients.

1. Methods

This was a retrospective, cohort study of infants aged 3-18 months seen at the Department of Pediatrics, Dana-Dwek Children’s Hospital, Sourasky Medical Center from March 2018 through March 2020. The study was approved by the hospital Ethics Review Board. Electronic

https://doi.org/10.1016/j.ajem.2022.04.011 0735-6757/(C) 2022 Published by Elsevier Inc.

medical records of infants presenting to the pediatric ED with a rectal temperature of >=38.0 ?C and with a bulging fontanelle were examined.

Table 1

Demographic and clinical details of study patients

We excluded infants <3 months of Age SInce it is our policy to routinely	N = 40
perform LP in this group of patients, and patients >18 months of age,	Age (months):
since the anterior fontanelle is usually closed by this age. We also ex- Mean		5.87
cluded patients with known immune compromise. Most of our patients Median		6
are fully, or at least partially, immunized against Haemophilus influenzae Range		3-13
type B (Hib) and Streptococcus pneumonia, although Neisseria vaccines still have to be purchased privately in our jurisdiction.	Gender, male (%) Immunization status?, number (%): Full	22 (55) 34 (85)

The following data were extracted onto an Excel (2007) spread- sheet: history, age, sex, presenting features, examination including vital signs, general appearance, complete blood count, C-reactive pro- tein levels, sodium level, blood culture results, cerebrospinal fluid (CSF) analysis results: white blood cell count and differential, Bacterial culture and Polymerase chain reaction results, and whether patients were admitted or discharged.

Long term outcomes were assessed by examining hospital medical records for two months after initial presentation to document outcomes of any return visits by our patients. Our center is a referral center serving our population, making it unlikely that sick patients would present, or be referred, to any center other than ours.

• 1. Definitions

Bulging fontanelle: A bulging anterior fontanelle with the head posi- tioned at 30-45? to the clavicle with no crying, vomiting, or coughing in the previous five minutes.

Meningitis: CSF WBC count >5 cells/uL CSF. Bacterial meningitis: bacterial growth in CSF culture.

Aseptic meningitis: WBC count >5 cells/uL CSF with negative CSF bacterial cultures, negative CSF Gram stain, and negative PCR (if done), in the absence of prior treatment with antimicrobial agents.

Severe bacterial infection [9]: Pneumonia, meningitis, septicemia, urinary tract infection, cellulitis, abscess, bacterial gastroenteritis, acute mastoiditis, lymphadenitis, bacteremia, Septic arthritis, or osteo- myelitis.

Pneumonia: New air-space opacities in the setting of acute respira- tory symptoms.

Well-appearing: normal feeding and appetite, normal activity (not lethargic, no decreased response to stimuli, not irritable or inconsolable, smiling (even if only after defervescence), normal tone, normal color (not mottled, cyanotic, or pale), normal respiratory effort, vital signs within normal limits for age.

• 1. Data analysis

Descriptive statistics were used to describe the study population. Categorical variables were described using frequency and percentage. Continuous variables with normal distribution were compared using the t-test, and those with non-normal distribution using the Wilcoxon test, and for categorical data, the Fisher exact test was used. P values of <0.05 were considered significant.

1. Results

Forty children were included in the study; their demographic, clini- cal, laboratory characteristics, diagnosis, and outcomes are shown in Table 1. Ages ranged from 3 to 13 months, with a mean of 5.9 months, and 22 (55%) were male. Most of our patients were fully, or at least par- tially, immunized against H. influenzae type B (Hib) and Streptococcus pneumoniae. No discharged patients returned and/or received an alter- native diagnosis, and there was no mortality. Thirteen patients (32.5%) underwent LP. Their clinical and laboratory data are compared to the 27 patients who did not undergo LP in Table 2. Well-looking infants were less likely (p = 0.028), and restless children with poor appetite more likely (p = 0.028 and 0.007 respectively) to undergo LP. The

Mortality 0

Partial	5 (12.5)
Not vaccinated	1 (2.5)
Temperature (?C): Mean Median	38.8 39
Range	38-41
White blood cell count (x1000/mm3):
Mean	10.98
Median	10.85
Range	4.6-24.0
C-reactive protein (mg/L):
Mean	20.6
Median	14
Range	0.01-136
Blood culture sent, number (%)	27 (67.5)
CSF culture sent, number (%)	13 (32.5)
Patients with serious bacterial infection, number (%):	2 (5)
Meningitis	2
Patients without serious bacterial infection, number (%):	38 (95)
Aseptic meningitis	1 (2.5)
Otitis media	1 (2.5)
Upper respiratory tract infection	1 (2.5)
Gastroenteritis	1 (2.5)
Urinary tract infection	1 (2.5)
Hydrocephalus	1 (2.5)
Viral infection	32 (80)
Outcomes, number (%):
Hospitalization	8 (20)
Ward	8 (87.5)
pediatric intensive care unit	1 (12.5)

* Immunization status against H. influenzae type B (Hib) and nine serotype pneumo- coccal (PCV9).

only clinical or laboratory result which correlated with meningitis (Table 3) was C-reactive protein (p = 0.0002). One patient had aseptic meningitis and two had bacterial meningitis. The two infants with bac- terial meningitis were both ill-appearing on presentation. The first was a 10-month-old who presented with vomiting and lethargy. LP results revealed pleocytosis with a WBC count of 229 cells/uL, and CSF and blood cultures were positive for S. pneumoniae. The second patient was an eight-month-old baby who presented with lethargy alone; CSF WBC count was 342 cells/uL, and CSF and blood cultures were positive for S. pneumoniae.

Table 2

Clinical and laboratory data of patients who did and did not undergo lumbar puncture (N = 40)

	LP (N = 13)	No LP (N = 27)	P value
Clinical presentation, number (%): Vomiting	1 (7.6%)	1 (3.7%)	0.45
Loss of appetite	3 (23.1%)	1 (3.7%)	0.084
Well looking	10 (76.9%)	27 (100%)	0.028
Restlessness	3 (23.1%)	0	0.028
Upper respiratory tract/Viral infection	8 (61.5%)	23 (85.1%)	0.082
Examined by resident only	8 (61.5%)	15 (44.4%)	0.16
Laboratory data, median: Hemoglobin (g/dL)	11.1	11.7	0.57
3	11.54	10.4	0.23
Platelets (x1000/mm3)	398	278	0.073
C-reactive protein (mg/L)	31.09	11.12	0.33
Sodium (mmol/L)	136.7	136.9	0.12
Negative blood culture	11/13 (79%)	14/14 (100%)	0.22

White blood cells (x1000/mm )

Table 3

Clinical and laboratory data of patients with and without meningitis (N = 40)

bulging fontanelle probably has a low risk of bacterial meningitis. Two studies, performed in our country, are of relevance but differ in that all

Meningitis (N = 2)

Laboratory tests, median
Hemoglobin (g/dL)	9.3	11.4364	0.413
White blood cell count (x1000/mm	3	11.13	0.436
Platelet count (x1000/mm3)	591.5	363.6	0.523
Sodium (mmol/L)	136.5	136.7	0.906
C-reactive protein (mg/L)	136	15.4	0.0002
Negative blood culture	0	25 (100%)	0.002

No meningitis (N = 38)

P

value

patients underwent LP. Takagi et al. [3] showed that of 764 infants, aged 2-18 months who underwent LP over a period of 10 years, 304 pre-

Clinical presentation Vomiting	0	2 (5.2%)	0.9	sented with a bulging fontanelle, and none of the infants who was de- scribed as well-appearing had bacterial meningitis. Moreover, of the
Well looking	0	38 (100%)	0.001	10 infants diagnosed with bacterial meningitis, only one presented

1. Discussion

) 13.8

with a bulging fontanelle. Shacham et al. described 153 infants with a bulging fontanelle, and none of those who was well appearing was among diagnosed with bacterial meningitis [4]. Limitations: This was a retrospective study conducted in a single Tertiary medical center in an urban setting, and the rate of bacterial meningitis in our well- vaccinated population was low, with only two patients having bacterial meningitis. Our study, combined with previous works, supports the rec- ommendation that management of well-appearing, fully vaccinated, in- fants, with normal medical history who present with fever and bulging fontanelle and normal vital signs can be managed without a LP.

Two of 40 febrile infants with a bulging fontanelle in our study pop- ulation were found to have bacterial meningitis. Both were distinguish- able on clinical grounds in terms of not being ‘well-looking’ or, exhibiting ‘restlessness’. The only laboratory measurement which helped distinguish these patients was CRP level. There were no ‘missed cases’ of meningitis. The prognosis of bacterial meningitis is critically dependent on a rapid causal diagnosis [10], and the institution of appro- priate antibiotic treatment and management [11]. For these reasons, physicians, especially those on the frontlines in pediatric ED’s, have a low threshold for performing a LP. Although a bulging fontanelle in a fe- brile infant is a possible sign of bacterial meningitis, and one textbook suggests that it occurs late in the course of the disease [12], seasoned cli- nicians report that this may not be the case (personal communication), creating uncertainty as to its true time course. Infants with bacterial meningitis are more likely to manifest vomiting, poor feeding, or leth- argy before presenting with a bulging fontanelle, hence the importance of these findings. Aseptic meningitis is more common than bacterial meningitis but its symptoms are usually less severe. Serious complica- tions are rare and most patients recover within two weeks. Bacterial meningitis carries a high mortality and morbidity [13,14], and the deci- sion to not perform a LP in a febrile infant with a bulging fontanelle may be controversial. In our study, this course of action was reserved only for well-looking children, and was never applied when there was any con- cern regarding the patient’s clinical status. Exclusion of meningitis in in- fants can be very difficult, even for an experienced physician, but for febrile, well-appearing, infants who are smiling and well-looking, with- out signs of restlessness, a bulging fontanelle appears to have a very low positive predictive value for meningitis [5]. Current protocols [3] sug- gest that if the physician decides on performing a LP in a febrile infant with a bulging fontanelle, the patient should be hospitalized and treated with antibiotics until the CSF culture results are available, even though this may result in increased antibiotic adverse events, nosocomial infec- tions, and risk of antibiotic resistance [6]. Most bacterial causes of men- ingitis in children are now preventable through immunization, making our findings of particular relevance to well-vaccinated populations. Our results suggest that a well-looking, fully vaccinated, febrile infant with a

CRediT authorship contribution statement

Jiriys Shahada: Investigation. Oren Tavor: Methodology. Or Segev: Writing - original draft. Ayelet Rimon: Writing - original draft. Dennis Scolnik: Writing - review & editing. Miguel Glatstein: Writing - review & editing, Writing - original draft, Formal analysis, Data curation.

References

1. Dery MA, Hasbun R. Changing epidemiology of bacterial meningitis. Curr Infect Dis Rep. 2007;9(4):301-7.
2. Van Ettekoven CN, van de Beek D, Brouwer MC. Update on community-acquired bacterial meningitis: guidance and challenges. Clin Microbiol Infect. 2017;23(9): 601-6.
3. Takagi D, Oren-Ziv A, Shles A, Schujovitzky D, Yechiam H, Rosenbloom E. Bulging fontanelle in febrile infants as a predictor of bacterial meningitis. Eur J Pediatr. 2021;180(4):1243-8.
4. Shacham S, Kozer E, Bahat H, Mordish Y, Goldman M. Bulging fontanelle in febrile infants: is lumbar puncture mandatory? Arch Dis Child. 2009;94(9):690-2.
5. Akpede GO, Abiodun PO, Ambe JP, Jacob DD. Presenting features of bacterial menin- gitis in young infants. Ann Trop Paediatr. 1994;14(3):245-52.
6. Beri S, Hussain N. Bulging fontanelle in febrile infants: lumbar puncture is manda- tory. Arch Dis Child. 2011;96(1):109.
7. Tamer SK, Tamer U, Warey P. Infantile pseudotumor cerebri related to Viral illness. Indian J Pediatr. 1996;63(5):645-9.
8. Mann NP, McLellan NJ, Cartlidge PH. Transient intracranial hypertension of infancy. Arch Dis Child. 1988;63(8):966-8.
9. Brauner M, Goldman M, Kozer E. Extreme leucocytosis and the risk of serious bacte- rial infections in febrile children. Arch Dis Child. 2010;95(3):209-12.
10. Bodilsen J, Brandt CT, Sharew A, Dalager-Pedersen M, Benfield T, Schonheyder HC, et al. Early versus late diagnosis in community-acquired bacterial meningitis: a ret- rospective cohort study. Clin Microbiol Infect. 2018;24(2):166-70.
11. Bodilsen J, Dalager-Pedersen M, Schonheyder HC, Nielsen H. Time to antibiotic ther- apy and outcome in bacterial meningitis: a Danish population-based cohort study. BMC Infect Dis. 2016;16:392.
12. Bachur RG, Shaw KN. Fleisher and Ludwig’s textbook of pediatric emergency medi- cine. Fleisher and Ludwig’s Textbook of Pediatric Emergency Medicine; 2015. p. 841.
13. Posadas E, Fisher J. Pediatric bacterial meningitis: an update on early identification and management. Pediatr Emerg Med Pract. 2018;15(11):1-20.
14. Davis S, Feikin D, Johnson HL. The effect of Haemophilus influenzae type B and pneu- mococcal conjugate vaccines on childhood meningitis mortality: a systematic review. BMC Public Health. 2013;13 Suppl 3(Suppl. 3):S21.