See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/279308087

Implementation of a Meningitis Care Bundle in

the Emergency Room Reduces Mortality
Associated With Acute Bacterial Meningitis

Article in Annals of Pharmacotherapy June 2015

Impact Factor: 2.06 DOI: 10.1177/1060028015586012 Source: PubMed

CITATION READS

1 128

12 authors, including:

Luigia Scudeller Rodolfo Sbrojavacca

Policlinico San Matteo Pavia Fondazione IR Azienda Ospedaliera Santa Maria della Mise
111 PUBLICATIONS 987 CITATIONS 19 PUBLICATIONS 198 CITATIONS

SEE PROFILE SEE PROFILE

Maddalena Giannella
University of Bologna
91 PUBLICATIONS 1,119 CITATIONS

SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Luigia Scudeller
letting you access and read them immediately. Retrieved on: 14 July 2016
586012
research-article2015
AOPXXX10.1177/1060028015586012Annals of PharmacotherapyViale et al

Research Report
Annals of Pharmacotherapy

Implementation of a Meningitis Care Bundle

18
The Author(s) 2015
Reprints and permissions:
in the Emergency Room Reduces Mortality sagepub.com/journalsPermissions.nav
DOI: 10.1177/1060028015586012

Associated With Acute Bacterial Meningitis aop.sagepub.com

Pierluigi Viale, MD1, Luigia Scudeller, MD2, Federico Pea, MD3,

Sara Tedeschi, MD1, Russell Lewis, PharmD, PhD1, Michele Bartoletti, MD1,
Rodolfo Sbrojavacca, MD4, Francesco Cristini, MD1, Fabio Tumietto, MD1,
Nicoletta Di Lauria, MD1, Giovanni Fasulo, MD1,
and Maddalena Giannella, MD, PhD1

Abstract
Background: Prompt administration of antibiotics, adjunctive steroid therapy, and optimization of antibiotic delivery
to cerebrospinal fluid (CSF) are factors associated with improved outcome of patients hospitalized for acute bacterial
meningitis (ABM). However, the impact of a bundle of these procedures has not been reported. Objective: To assess
mortality and neurological sequelae at hospital discharge in a cohort of patients with ABM managed according to a
predefined bundle. Methods: Prospective study of all the patients hospitalized for ABM in two provinces of Northern Italy,
over two consecutive periods (2005-2009, 2010-2013). The bundle included: i) supportive care if needed; ii) immediate
administration of dexamethasone and 3rd generation cephalosporin; and iii) addition of levofloxacin if turbid CSF. Patients
managed according to the bundle were compared with a historical group of patients cared for ABM before the bundle was
implemented. Results: Overall, 85 patients with ABM were managed according to the bundle and were compared with 92
historical controls. In-hospital mortality rates for bundle and control group were 4.7% and 14.1% (p=.04). Among survivors,
13.5% and 18.9% (p=.4) of bundle and control-group patients presented neurological sequelae. The only variable associated
with mortality at multivariate analysis was ICU admission (HR 3.65). After adjusting for ICU admission, patients managed
according with the ABM bundle had significantly lower mortality rate compared to historical controls. Conclusions: Use
of a bundled protocol and antibiotics with excellent CSF penetration for the initial management of ABM in emergency
department is feasible and associated with significant reduction in mortality.

Introduction (BBB).7-9 The main antibiotics currently recommended by

Acute community-acquired bacterial meningitis is a life-
threatening disease associated with substantial mortality
and morbidity. In developed countries, the incidence of
meningitis has decreased over time owing to vaccination
programs against the leading causative pathogens.1
However, the case fatality rate has not changed significantly
and ranges from 7% to 30% for episodes caused by Neisseria
meningitidis and Streptococcus pneumoniae, respectively.1-3
Neurological sequelae, including hearing loss, develop-
mental disorders, and neuropsychological impairment,
occur in up to 50% of disease survivors.1,3-5
Prompt diagnosis, early administration of dexametha-
sone, early appropriate antibiotic therapy, and supportive
treatment have been shown to be key factors that improve
the prognosis of acute bacterial meningitis (ABM).6
However, the concept of appropriate antimicrobial therapy
has undergone some reconsideration in recent years mainly
in terms of penetration across the blood-brain barrier
the guidelines for treatment of ABM are -lactams and van-
comycin, which may have suboptimal penetration across
the BBB.10 Accordingly, some authors suggested as a pos-
sible solution the addition of an anti-Gram-positive fluoro-
quinolone in regions where pneumococcal resistance is low
1
Infectious Diseases Unit - Department of Medical and Surgical Sciences,
University of Bologna, Teaching Hospital S. Orsola-Malpighi Bologna,
Italy
2
Clinical Epidemiology and Biostatistics Unit, Scientific Direction, IRCCS
Policlinic San Matteo Foundation, Pavia, Italy
3
Institute of Clinical Pharmacology & Toxicology, Azienda Ospedaliero-
Universitaria Santa Maria della Misericordia, Udine, Department of
Experimental and Clinical Medical Sciences, University of Udine, Udine,
Italy
4
Department of Medicine, Azienda Ospedaliero-Universitaria Santa
Maria della Misericordia, Udine, Italy
Corresponding Author:
Maddalena Giannella, Via Masserenti 11, Bologna, 40137, Italy.
Email: maddalena.giannella@libero.it

Downloaded from aop.sagepub.com at Policlinico San Matteo on June 24, 2015

2 Annals of Pharmacotherapy

Figure 1. Flow chart of the meningitis bundle.

Abbreviations: CSF, cerebrospinal fluid; ED, emergency department; LD, loading dose; LP, lumbar puncture.

because of the excellent penetration of the drug into the
cerebrospinal fluid (CSF), irrespective of the BBB status.9
On this basis, we implemented a meningitis bundle,
with interventions starting from the ABM clinical suspicion
in the emergency department (ED) and arriving to the anti-
bacterial choice in order to optimize the approach to the
patients with ABM and to improve their clinical outcome.
The aim of this study was to assess the clinical outcome of all
the consecutive patients with ABM treated according to the
meningitis bundle in 2 provinces of northern Italy, comparing
data with a historical control group of patients with ABM
who were hospitalized before the bundle was implemented.
Material and Methods
Study Design and Setting
We prospectively studied all the consecutive adult patients
with ABM managed according to a preestablished bundle to
assess the rates of in-hospital mortality and neurological
sequelae at hospital discharge. Secondary outcomes were
time to clinical stability and length of hospital stay. Patients
managed according to the bundle were compared with a his-
torical control group.
The bundle was implemented in the hospitals of 2 prov-
inces of 2 regions of northern Italy over 2 subsequent time
periods (Udine, from January 2005 to December 2009;
Bologna, from January 2010 until today). For this study, we
analyzed data up to December 2013.
The province of Udine covers an area of 4905 km2 and a
total population of about 0.5 million, with a teaching ter-
tiary referral hospital of 950 beds and 5 secondary hospi-
tals. The province of Bologna covers an area of 3702 km2
and a total population of about 1 million, with a teaching
tertiary referral hospital of 1420 beds and 7 secondary hos-
pitals. The study was approved by the local ethics commit-
tees in both regions.
The meningitis bundle is depicted in Figure 1. The
bundle was developed by a multidisciplinary team of spe-
cialists (infectious disease, emergency, intensive care, clini-
cal pharmacology, and clinical microbiology). Briefly, the
bundle consisted of the following:
1. immediate supportive care if needed;
2. administration within 1 hour of initial patient evalu-
ation (before lumbar puncture [LP] if necessary) 10
mg of dexamethasone and a dose of a third- genera-
tion cephalosporin (cefotaxime 4 g or ceftriaxone

Downloaded from aop.sagepub.com at Policlinico San Matteo on June 24, 2015

Viale et al 3
2 g)11; after the first 10-mg dose, dexamethasone
was administered in all patients at the dosage of
0.15 mg/kg every 6 hours for 96 hours12;
3. if LP yielded a turbid (cloudy) CSF, the patient was
also administered levofloxacin (500 mg every 12
hours if creatinine clearance was >120 mL/min, 750
mg every 24 hours if creatinine clearance was
between 50 and 120 mL/min, and 500 mg every 24
to 48 hours if creatinine clearance was <50 mL/
min), in addition to the third-generation cephalospo-
rin; and
4. when patients were clinically stable and recovered
neurologically, antibiotic treatment was shifted to
oral levofloxacin monotherapy.
During the meningitis bundle period, all adult patients
presenting with a suspicion of meningitis to the ED of a
secondary participant hospital were transferred to the ED of
the referral tertiary teaching hospital after carrying out the
first 2 bundle steps. To ensure a high adherence to the bun-
dle over time and owing to the fast turnover of ED person-
nel, a periodic training about immediate administration of
therapy, microbiological workup, and overall management
was performed, and a graphic of the study flow chart sum-
marizing the main points of the bundle was provided to all
the EDs of the study setting.
Study Population
All adult (age 18 years) patients who presented to the ED
with a meningitis syndrome, defined by the presence of at
least 2 of the following, were included: fever (body tem-
perature 38C), neck stiffness, headache, and changes in
mental status.13 The diagnosis of ABM was then confirmed
by CSF examination showing the following: CSF-blood
glucose ratio <0.23, CSF protein concentration >220 mg/
dL, CSF leukocyte count >2000/mm3.13 Patients were
excluded from analysis if bacterial meningitis was not
confirmed.
The historical controls were all the consecutive adult
patients directly admitted or referred to the teaching hospi-
tal of Bologna with an ABM diagnosis from 2002 to 2009.
Variables
The following data were recorded: age, gender, and prior
antibiotic exposure within 7 days before ED presentation
and in ED prior to LP (a combined variable was also pro-
duced for antibiotic drug prior to LP). For clinical presen-
tation, we recorded body temperature, systolic and diastolic
blood pressure, heart rate, respiratory rate, and mental sta-
tus; APACHE II score was also calculated.14 Laboratory
data were collected for each patient, including CSF leuko-
cyte count, glucose and protein, peripheral leukocyte count,
Downloaded from aop.sagepub.com at Policlinico San Matteo on June 24, 2015
hemoglobin level, C-reactive protein, creatinine, lactate,
total protein and glucose, and microbiological results of
CSF culture and blood culture (on all CSF samples, the anti-
gen tests for S pneumoniae and N meningitidis were also
performed, but the results were recorded only when the
CSF culture was negative). For antibiotic therapy, the drug,
dose, and method of administration were recorded daily
throughout the hospital stay.
Outcomes were assessed at discharge and 30 days after
diagnosis. During the bundle period, patients were also
assessed after discharge during a follow-up outpatient visit
within 30 to 40 days of diagnosis.
Statistical Methods
We estimated that 80 patients per study group (n = 160
total) would be required to detect a mortality difference in
this pilot trial with 95% confidence that the 2-sided CI for
mortality would be between 7% and 23%.15
Descriptive statistics were produced for demographic,
clinical, and laboratory characteristics of cases. Means and
SDs are presented for normally distributed variables, medi-
ans and interquartile range (IQR) for nonnormally distrib-
uted variables, and numbers and percentages for categorical
variables. Of note, for the historical control group, complete
data on demographic, microbiological, therapy, and out-
come data were obtained (ie, data on vital signs or on labo-
ratory tests at admission were not always retrievable).
Groups were compared using parametric or nonparamet-
ric tests, according to data distribution, for continuous vari-
ables, and using Pearsons 2 test (Fishers exact test when
appropriate) for categorical variables. Time on study was
calculated in days from admission and censored at dis-
charge (for patients dying on the day of admission, 0.5 day
was imputed). Survival was estimated by the Kaplan-Meier
method; group comparison was assessed by the log-rank
test; and the association of a number of factors with mortal-
ity (both overall and within 30 days) and with time to stabil-
ity was explored using univariate and multivariate Cox
regression models. The likelihood ratio test was used to
assess relevance of the variables included, with P < 0.05 as
cutoff. Stata computer software version 13 (Stata
Corporation, 4905 Lakeway Drive, College Station, TX)
was used for statistical analysis.
Results
During the study period, approximately 850 patients were
evaluated in the ED of the participating hospitals for suspi-
cion of ABM. ABM diagnosis was confirmed in 89 patients.
Of these, 85 (95.5%) were treated according to the bundle,
40 (47.1%) from Udine and 45 (52.9%) from Bologna. The
reasons for exclusion of the 4 patients were the following:
(1) history of fluoroquinolone allergy, (2) delayed diagnosis
4 Annals of Pharmacotherapy

Table 1. Comparison of Demographic, Clinical, and Laboratory Data at Presentation.

Historical Group, n = 92 (%) Bundle Group, n = 85 (%) P

Demographic data
Age (years), mean SD 50.7 19.2 53.3 18 0.30
Male 50 (54.3) 44 (52) 0.76
Transferred from a secondary hospital 39 (42.4) 43 (50.5) 0.36
Ward of first admission
Intensive care unit 20 (21.7) 31 (36.5) 0.03
Infectious diseases 72 (78.3) 54 (63.5)
Clinical presentation at ED admission
Altered mental status 74 (80.4) 71 (83.5) 0.69
Systolic BPa 130 (120-150) 127 (110-140) 0.34
Diastolic BPa 80 (60-80) 70 (60-80) 0.13
Heart rate 90 (80-104)
RR 18 (15-22)
Body temperaturea 38.8 (38.2-39.2) 38.1 (37.3-38.9) 0.09
Number of SIRS parametersa 2 (1-3) 3 (2-5) <0.001
APACHE II score 12 (8-16) 12 (9-14) 0.42
Peripheral laboratory findings
Leukocytes (cells/mm3)a 18 690 (12 630-24 580) 16 185 (11 780-21 600) 0.06
Glucose (mg/dL)a 142 (114-185) 147 (130-171) 0.44
Lactate (mmol/L)a 2.1 (1.4-3.4)
C-reactive protein (mg/dL)a 19 (11.9-31) 15 (3.4-28.5) 0.02
Creatinine (mg/dL)a 1 (0.8-1.3) 0.92 (0.75-1.08) 0.05
CSF laboratory findings
Leukocytes (cells/mm3)a 1715 (482-4713.5) 1300 (425-3452) 0.37
Glucose (mg/dL)a 10 (3-34) 12.5 (3-44) 0.45
Protein (mg/dL)a 411 (274.5-632.5) 583 (231-2670) 0.05
CSF-Blood glucose ratioa 0.06 (0.03-0.22) 0.10 (0.02-0.30) 0.42

Abbreviations: BP, blood pressure; CSF, cerebrospinal fluid; ED, emergency department; RR, respiratory rate; SIRS, systemic inflammatory response
syndrome.
a
Continuous variables expressed as median and interquartile range.

of bacterial meningitis (after 36 hours from the ED admis-
sion because of suspicion for a cerebrovascular event), (3)
appearance of a skin rash within 48 hours of starting on
antibiotic therapy, and (4) development of ventilator-associ-
ated pneumonia within 72 hours of intensive care unit (ICU)
admission, requiring changes in the antimicrobial regimen.
The 85 patients managed according to the bundle were
compared with 92 historical controls. Demographic data
and clinical and laboratory parameters at presentation are
shown in Table 1. The bundle group patients were more fre-
quently admitted to the ICU and presented with a higher
number of systemic inflammatory response syndrome
(SIRS) parameters, whereas the historical controls had a
higher median value of C-reactive protein at the time of
presentation.
Use of Microbiological Resources and Etiology
Microbiological workup for ABM was performed in 100%
of patients. As shown in Table 2, blood cultures were drawn
more frequently in the bundle group than in the historical
controls. Compared with historical controls, patients man-
aged according to the ABM bundle were almost twice as
likely to have a pathogen isolated from CSF (odds ratio =
1.8; 95% CI = 0.99-3.32; P = 0.05) despite a higher fre-
quency of being exposed to a cephalosporin dose before LP.
The leading pathogens were S pneumoniae, N meningitides,
and Listeria monocytogenes in both groups.
Therapeutic Management
Patients in the bundle group received the first dose of anti-
biotics within 1 hour from arrival to the ED in 100% of
cases as compared with 25% of control patients (P < 0.001;
see Table 2). Among historical controls, 12 patients received
ceftriaxone monotherapy (2 g every 12 hours), whereas 80
patients received combination therapy based on ceftriaxone
(2 g every 12 hours) plus ampicillin (3 g every 6 hours),
vancomycin (1 g every 12 hours), or chloramphenicol (1 g
every 6 hours) in 70%, 22.5%, and 7.5% of cases, respec-
tively. The median days of combination therapy were 12
and 7 for the historical and bundle groups, respectively.

Downloaded from aop.sagepub.com at Policlinico San Matteo on June 24, 2015

Viale et al 5

Table 2. Microbiological Workup, Etiology, Therapeutic Management, and Outcome.

Historical Group, n = 92 (%) Bundle Group, n = 85 (%) P

Lumbar puncture
Performed 89 (96.7) 85 (100) 0.89
CSF culture positive 45 (48.9) 54 (63.5) 0.05
Antigen test positive 13 (14.1) 9 (28) 0.50
Blood culture
Performed 34 (37) 49 (57.6) 0.007
Positive 14 (15.2) 25 (29.4) 0.03
Etiological diagnosis
Yes 67 (72.8) 65 (76.5) 0.61
Causative agent
Streptococcus pneumoniae 40 (43.5) 40 (47.1) 0.56
Neisseria meningitidis 18 (19.6) 13 (15.3)
Haemophilus influenzae 0 3 (3.5)
Listeria monocytogenes 5 (5.4) 6 (7.1)
Other 4 (4.3) 3 (3.5)
Therapeutic management
Timing to first antibiotic dose <0.001
1 hour 23 (25) 85 (100)
>1-3 hours 27 (29.3) 0
>3 hours 42 (45.7) 0
Cephalosporin high dose prior to LP 10 (10.9) 43 (50.6) <0.001
Any antibiotic before LP 26 (28.2) 53 (62.4) <0.001
Dexamethasone administration 31 (33.7) 85 (100) <0.001
Days of combination therapya 12 (7-15) 8 (6-12) 0.002
Days of antibiotic therapya 14 (10-18) 14 (10-17) 0.51
Outcome
In-hospital mortality 13 (14.1) 4 (4.7) 0.04
Neurological sequelae at discharge 15/79 (18.9) 11/81 (13.5) 0.40
Days to clinical stabilitya 3 (2-5) 4 (3-6) 0.34
Days of hospital staya 15 (10-20) 13 (9-18) 0.28

Abbreviations: CSF, cerebrospinal fluid; ED, emergency department; LP, lumbar puncture.
a
Continuous variables expressed as medians and interquartile range.

Outcomes and Risk Factors for In-hospital

Mortality
The cumulative follow-up time was 1605.5 patient days for
bundle patients and 1721.5 patient days for historical controls
(median = 15.5 vs 13 days, P = 0.19). Among the bundle
patients, in-hospital mortality rate was 4.7% (4/85): 3 patients
died in the ICU within 1, 5, and 15 days from diagnosis,
respectively; the remaining patient died in a medical ward
within 43 days of diagnosis, with hairy-cell leukemia and
invasive aspergillosis. Among controls, 13 died in the hospital
(14.1%). Difference in 30-day survival between bundle and
historical patients was statistically significant (P = 0.005;
Figure 2).
Out of 81 survivors, 11 (13.5%) had neurological sequelae
at discharge (median = 13 days to discharge): cranial nerve Figure 2. Kaplan-Meier estimates of survival in bundle patients
palsies (n = 4), hearing impairment (n = 3), focal neurological and historical controls.
deficit secondary to postmeningitis cerebrovascular event *Adjusted for intensive care unit admission.

Downloaded from aop.sagepub.com at Policlinico San Matteo on June 24, 2015

6 Annals of Pharmacotherapy

Table 3. Univariate and Multivariate Cox Regression Analysis of Risk Factors Associated With Mortality Within 30 Days.

Univariate Analysis Multivariate Analysis

HR (95% CI) P HR (95%CI) P

Age 1.01 (0.98-1.03) 0.57
Female sex 0.82 (0.31-2.21) 0.69
ICU admission 2.96 (1.1-8) 0.03 3.65 (1.32-10.08) 0.01
Number of SIRS parameters 1.13 (0.83-1.54) 0.42
Altered mental status Not calculated 0.06a
Timing to the first antibiotic dose
1 hour 1.00 (base)
>1-3 hours 1.62 (0.40-6.54) 0.50
>3 hours 0.79 (0.18-3.53) 0.75
Dexamethasone administration 0.34 (0.12-0.94) 0.04
Any antibiotic before LP 0.61 (0.23-1.77) 0.37
Etiological diagnosis 1.28 (0.36-4.52) 0.69
Streptococcus pneumoniae 0.89 (0.34-2.33) 0.81
Neisseria meningitidis 1.16 (0.33-4.06) 0.81
Bundle group 0.27 (0.76-0.95) 0.04 0.22 (0.04-1.13) 0.07

Abbreviations: CSF, cerebrospinal fluid; HR, hazard ratio; LP, lumbar puncture; SIRS, systemic inflammatory response syndrome.
a
None of the patients with normal mental status at diagnosis died; P value is obtained from log-rank test rather than Cox regression.

(n = 2), tinnitus (n = 1), and coma (minimally conscious state,
n = 1). At 30 days from ABM diagnosis, all patients with
cerebrovascular events fully recovered. Patients with cranial
nerve palsies improved, but deficits were still present, and the
patients with other sequelae were stable.
Neurological sequelae at discharge were present in 15/79
(18.9%) of controls (P = 0.40 when compared with bundle
patients, at a median of 15 days to discharge). Follow-up
data for this group were missing.
Median times to clinical stability were 4 (IQR = 3-6) and
3 (IQR = 2-5) days (P = 0.35) between bundle and control
patients, respectively. Median lengths of hospital stay were
13 (IQR = 9-18) and 15 (IQR = 10-20) days (P = 0.19) for
bundle and control groups, respectively.
Univariate Cox regression analysis of risk factors for in-
hospital mortality showed that the need for ICU admission
was associated with higher mortality, whereas dexametha-
sone administration and management according to the care
bundle were factors associated with better outcome (Table 3).
On multivariate analysis, ICU admission retained statistical
significance (Table 3). Analysis of survival in bundle and
control groups showed a significantly higher survival rate
among bundle patients also after adjusting the analysis for
severity (ICU admission; Figure 2).
Discussion
Rates of bacterial meningitis have decreased in the past 2
decades, but the infection continues to be associated with high
morbidity and mortality. In a recent series of 3188 patients
with bacterial meningitis observed during 1998-2007 in North
America, the case fatality rate did not change significantly,
varying between 15.7% in 1998-1999 and 14.3% in 2006-
2007 (P = 0.50).1 Although reasons for the persistently high
mortality are not entirely understood, prompt administration
of antibiotics, adjunctive steroid therapy, and optimization of
antibiotic delivery to CSF are factors known to improve
patient outcomes. In this historical case control study, we
found that a care bundle designed to optimize these treatment
variables in the ED was associated with a lower mortality rate
(4.7%) compared with that in historical controls with menin-
gitis syndromes not managed according to the care bundle
(14.1%).
Key elements of the bundle are the early supportive care
if necessary and the immediate administration of dexameth-
asone and a dose of a third-generation cephalosporin soon
after suspicion for ABM. Evidence supporting these strate-
gies is derived from clinical studies that demonstrated
favorable impact on mortality and reduction of neurological
sequelae among patients receiving early intensive support-
ive care, antibiotic therapy, and dexamethasone administra-
tion within the first hour of presentation with a meningitis
syndrome.2,12,16-18
Care bundles are increasingly recognized as effective
methods for improving care and survival, particularly (but
by no means exclusively) in emergency situations.19
However, there are well-recognized barriers to their appli-
cation.20 Implementation of a bundle with high reliability
requires a redesign of work processes, along with commu-
nication strategies, infrastructure, sustained measurement,
and regular feedback.19 Indeed, implementation of our men-
ingitis bundle required a great preparatory effort in the

Downloaded from aop.sagepub.com at Policlinico San Matteo on June 24, 2015

Viale et al 7
receiving area of the 2 participating hospitals to ensure
100% participation of all medical and nonmedical ED
workers, transportation staff, and Infectious Diseases (ID)
and ICU personnel, 24 hours a day, during the period of the
study. Furthermore, we showed that the early administra-
tion of antibiotics in the ED before LP was not associated
with a lower yield of CSF culture. Thus, it is reasonable to
administer the first dose of antibiotics before starting the
microbiological workup when immediate LP is not feasible.
This could be a key message for physicians working in
peripheral institutions with less resources for managing
critically ill patientssimply recognizing the infection and
administering dexamethasone and an antibiotic dose before
referring the patient to the tertiary hospital can reduce mor-
tality. In addition, when performed, blood cultures drawn
before antibiotic administration in our cohort improved the
diagnostic yield, confirming their key role in the diagnostic
workup. Indeed, bacteremia is more common in patients
with meningitis compared with other common community-
acquired infectious diseases (eg, pneumonia).21 Therefore,
immediate collection of blood cultures is now included as a
key component in the meningitis bundles currently fol-
lowed in our hospital.
Although the original steps of the bundle were followed
with great accuracy, we found that physicians frequently
administered longer courses of antibiotic therapy than rec-
ommended. The prolonged treatment course is consistent
with the belief among many physicians that serious infec-
tions should be treated for a prolonged period of time. In
our opinion, the duration of antibiotic therapy should be
based on the clinical stability and normalization of inflam-
matory markers.
Our care bundle also required the administration of levo-
floxacin if a patient was found to have a cloudy CSF follow-
ing LP. Fluoroquinolones have favorable PK behavior in the
central nervous system when administered systemically.
Levofloxacin, in particular, rapidly distributes into the cen-
tral nervous system and achieves relatively higher exposures
over time in CSF compared with vancomycin or many
-lactams (mean area under the curve [AUC]CSF/AUCserum
ratio of 0.7) in the absence of meningeal inflammation.
Therefore, the addition of levofloxacin may guarantee suf-
ficient levels of active drug even as inflammation decreases
at the BBB, reducing the risk of treatment relapse.22,23 In
experimental models of meningitis caused by penicillin-
resistant S pneumoniae, combinations of levofloxacin
-lactams are synergistic and reduced the emergence of
levofloxacin resistance.24,25 Furthermore, the resistance rates
to cephalosporins and fluoroquinolones for S pneumoniae in
Europe remain low: 0.3% and <3%, respectively.26,27 In a
recent cross-sectional study to assess the prevalence, sero-
type distribution, and antibiotic susceptibility of S pneu-
moniae isolates carried by healthy children in northern Italy,
96.5% and 100% of strains were susceptible to ceftriaxone
Downloaded from aop.sagepub.com at Policlinico San Matteo on June 24, 2015
and levofloxacin, respectively.28 Finally, fluoroquinolones
also have in vitro activity against L monocytogenes,29,30 and
interestingly, all the tested strains were found to be suscep-
tible when considering a minimal inhibitory concentration
(MIC) breakpoint for levofloxacin of <2 mg/L.28 In clinical
practice, levofloxacin has also been used successfully in the
treatment of L monocytogenes meningitis.31
Our study has some limitations. First, the use of a histori-
cal monocentric control group may be a limitation because
possible risk factors influencing outcome that were not ana-
lyzed may differ between historical and bundle groups.
Nevertheless, the 2 groups were comparable in terms of age,
sex, and comorbidities. Furthermore, in the prebundle era, it
was not mandatory to centralize patients with ABM to the
referral tertiary care hospital. Therefore, less-severe cases
may not have been included in the historical group even
though patients managed according to the bundle in our
series were more likely to be admitted to the ICU and have a
higher median number of SIRS criteria. Moreover, the dif-
ference in survival between historical and bundle patients
remained even after adjusting for ICU admission. Second,
data on MICs for cephalosporin and levofloxacin were
incomplete; therefore, we could not assess the adequacy of
therapy based on MICs. However, all available isolates from
CSF and blood cultures were fully susceptible to both antibi-
otics. Third, we did not collect data of patients with suspi-
cion of ABM in which the diagnosis was ruled out after
initial management according to the bundle. Therefore, the
incidence of adverse effects associated with initial cephalo-
sporin and/or steroid doses may be underestimated. Finally,
we did not perform cost-effectiveness analysis.
In conclusion, our experience shows that the application
of a meningitis bundle in ED is feasible and associated with
a reduction in the mortality rate in patients with ABM.
Future studies should investigate a role for fluoroquino-
lones in the treatment of ABM management.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of this
article.
Funding
The author(s) received no financial support for the research,
authorship, and/or publication of this article.
References
1. Thigpen MC, Whitney CG, Messonnier NE, et al. Bacterial
meningitis in the United States, 1998-2007. N Engl J Med.
2011;364:2016-2025.
2. Glimaker M, Johansson B, Grindborg O, Bottai M, Lindquist
L, Sjolin J. Adult bacterial meningitis: earlier treatment and
improved outcome following guideline revision promoting
prompt lumbar puncture. Clin Infect Dis. 2015;60:1162-1169.
8 Annals of Pharmacotherapy
3. van de Beek D, de Gans J, Spanjaard L, Weisfelt M, Reitsma
JB, Vermeulen M. Clinical features and prognostic fac-
tors in adults with bacterial meningitis. N Engl J Med.
2004;351:1849-1859.
4. van de Beek D, Farrar JJ, de Gans J, et al. Adjunctive dexa-
methasone in bacterial meningitis: a meta-analysis of indi-
vidual patient data. Lancet Neurol. 2010;9:254-263.
5. van de Beek D, de Gans J, Tunkel AR, Wijdicks EF.
Community-acquired bacterial meningitis in adults. N Engl J
Med. 2006;354:44-53.
6. van de Beek D, de Gans J, McIntyre P, Prasad K. Steroids in
adults with acute bacterial meningitis: a systematic review.
Lancet Infect Dis. 2004;4:139-143.
7. Pea F, Viale P. The antimicrobial therapy puzzle: could phar-
macokinetic-pharmacodynamic relationships be helpful in
addressing the issue of appropriate pneumonia treatment in
critically ill patients? Clin Infect Dis. 2006;42:1764-1771.
8. van de Beek D, Brouwer MC, Thwaites GE, Tunkel AR.
Advances in treatment of bacterial meningitis. Lancet.
2012;380:1693-1702.
9. Nau R, Djukic M, Spreer A, Eiffert H. Bacterial meningitis:
new therapeutic approaches. Expert Rev Anti Infect Ther.
2013;11:1079-1095.
10. Tunkel AR, Hartman BJ, Kaplan SL, et al. Practice guidelines
for the management of bacterial meningitis. Clin Infect Dis.
2004;39:1267-1284.
11. Udy AA, Varghese JM, Altukroni M, et al. Subtherapeutic
initial beta-lactam concentrations in select critically ill
patients: association between augmented renal clearance and
low trough drug concentrations. Chest. 2012;142:30-39.
12. de Gans J, van de Beek D. Dexamethasone in adults with bac-
terial meningitis. N Engl J Med. 2002;347:1549-1556.
13. van de Beek D. Progress and challenges in bacterial meningi-
tis. Lancet. 2012;380:1623-1624.
14. Knaus WA, Draper EA, Wagner DP, Zimmerman JE.
APACHE II: a severity of disease classification system. Crit
Care Med. 1985;13:818-829.
15. Cocks K, Torgerson DJ. Sample size calculations for pilot
randomized trials: a confidence interval approach. J Clin
Epidemiol. 2013;66:197-201.
16. Miner JR, Heegaard W, Mapes A, Biros M. Presentation,
time to antibiotics, and mortality of patients with bacterial
meningitis at an urban county medical center. J Emerg Med.
2001;21:387-392.
17. Proulx N, Frechette D, Toye B, Chan J, Kravcik S. Delays in
the administration of antibiotics are associated with mortality
from adult acute bacterial meningitis. QJM. 2005;98:291-298.
18. Glimaker M, Johansson B, Bell M, et al. Early lumbar punc-
ture in adult bacterial meningitis: rationale for revised guide-
lines. Scand J Infect Dis. 2013;45:657-663.
19. Stoneking L, Denninghoff K, Deluca L, Keim SM, Munger
B. Sepsis bundles and compliance with clinical guidelines. J
Intensive Care Med. 2011;26:172-182.
Downloaded from aop.sagepub.com at Policlinico San Matteo on June 24, 2015
20. Fong JJ, Cecere K, Unterborn J, Garpestad E, Klee M, Devlin
JW. Factors influencing variability in compliance rates and
clinical outcomes among three different severe sepsis bun-
dles. Ann Pharmacother. 2007;41:929-936.
21. Brouwer MC, Tunkel AR, van de Beek D. Epidemiology,
diagnosis, and antimicrobial treatment of acute bacterial men-
ingitis. Clin Microbiol Rev. 2010;23:467-492.
22. Pea F, Pavan F, Nascimben E, et al. Levofloxacin disposition
in cerebrospinal fluid in patients with external ventriculos-
tomy. Antimicrob Agents Chemother. 2003;47:3104-3108.
23. Nau R, Sorgel F, Eiffert H. Penetration of drugs through the
blood-cerebrospinal fluid/blood-brain barrier for treatment
of central nervous system infections. Clin Microbiol Rev.
2010;23:858-883.
24. Kuhn F, Cottagnoud M, Acosta F, Flatz L, Entenza J,
Cottagnoud P. Cefotaxime acts synergistically with
levofloxacin in experimental meningitis due to penicil-
lin-resistant pneumococci and prevents selection of levoflox-
acin-resistant mutants in vitro. Antimicrob Agents Chemother.
2003;47:2487-2491.
25. Flatz L, Cottagnoud M, Kuhn F, Entenza J, Stucki A,
Cottagnoud P. Ceftriaxone acts synergistically with levo-
floxacin in experimental meningitis and reduces levofloxa-
cin-induced resistance in penicillin-resistant pneumococci. J
Antimicrob Chemother. 2004;53:305-310.
26. de la Campa AG, Ardanuy C, Balsalobre L, et al. Changes
in fluoroquinolone-resistant Streptococcus pneumoniae after
7-valent conjugate vaccination, Spain. Emerg Infect Dis.
2009;15:905-911.
27. Reinert RR, Reinert S, van der Linden M, Cil MY,
Al-Lahham A, Appelbaum P. Antimicrobial susceptibility
of Streptococcus pneumoniae in eight European countries
from 2001 to 2003. Antimicrob Agents Chemother. 2005;49:
2903-2913.
28. Zuccotti G, Mameli C, Daprai L, et al. Serotype distribution
and antimicrobial susceptibilities of nasopharyngeal isolates
of Streptococcus pneumoniae from healthy children in the
13-valent pneumococcal conjugate vaccine era. Vaccine.
2014;32:527-534.
29. Martinez-Martinez L, Pascual A, Suarez AI, Perea EJ.
In-vitro activity of levofloxacin, ofloxacin and D-ofloxacin
against coryneform bacteria and Listeria monocytogenes. J
Antimicrob Chemother. 1999;43(suppl C):27-32.
30. Seral C, Barcia-Macay M, Mingeot-Leclercq MP, Tulkens
PM, Van Bambeke F. Comparative activity of quinolones
(ciprofloxacin, levofloxacin, moxifloxacin and garenoxacin)
against extracellular and intracellular infection by Listeria
monocytogenes and Staphylococcus aureus in J774 macro-
phages. J Antimicrob Chemother. 2005;55:511-517.
31. Viale P, Furlanut M, Cristini F, Cadeo B, Pavan F, Pea
F. Major role of levofloxacin in the treatment of a case of
Listeria monocytogenes meningitis. Diagn Microbiol Infect
Dis. 2007;58:137-139.