Acute Bacterial Meningitis in Qatar: A Hospital-Based Study from 2009 to 2013

Background and Objectives Bacterial meningitis is a common medical condition in Qatar. The aim of this study was to describe the clinical characteristics of bacterial meningitis, the frequency of each pathogen, and its sensitivity to antibiotics and risk factors for death. Patients and Methods This retrospective study was conducted at Hamad General Hospital between January 1, 2009, and December 31, 2013. Results We identified 117 episodes of acute bacterial meningitis in 110 patients. Their mean age was 26.4 ± 22.3 years (range: 2–74) and 81 (69.2%) of them were male patients. Fifty-nine episodes (50.4%) were community-acquired infection and fever was the most frequent symptom (94%), whereas neurosurgery is the most common underlying condition. Coagulase-negative staphylococci were the most common causative agent, of which 95% were oxacillin-resistant, while 63.3% of Acinetobacter spp. showed resistance to meropenem. The in-hospital mortality was 14 (12%). Only the presence of underlying diseases, hypotension, and inappropriate treatment were found to be independent predictors of mortality. Conclusion Acute bacterial meningitis predominantly affected adults and coagulase-negative staphylococci species were the common causative agent in Qatar with majority of infections occurring nosocomially. More than 90% of all implicated coagulase-negative staphylococci strains were oxacillin-resistant.


Introduction
Despite medical advances, acute bacterial meningitis (ABM) constitutes a global public health problem, especially in developing countries with poor health facilities due to high rates of malnutrition, poor living conditions, and lack of access to appropriate preventive and curative services that may predispose people to the disease and reduce their chances of receiving optimal treatment [1,2]. In developed countries, the burden of the disease has reduced and its epidemiology has changed as a result of the widespread use of vaccines against the most common meningeal pathogens [3].
Accurate information on important etiologic agents and populations at risk is needed to determine public health measures and ensure appropriate management of ABM [3]. In Qatar, although ABM is a common medical condition that physicians face, there are few reports describing this disease [4][5][6]. We conducted the present study, the purposes of which were to (1) describe the demographic and clinical characteristics of ABM, (2) determine the relative frequency of each pathogen and its susceptibility to various antimicrobial agents, and (3) determine the outcome and the significant predictors of the outcome among patients with ABM in Qatar.

Design and
Setting. This retrospective descriptive study, which involved all in-patients with ABM, was conducted at Hamad General Hospital between January 1, 2009, and December 31, 2013. This hospital is a 603-bed tertiary care center that covers all specialties except for hematologyoncology, cardiology, and obstetrics and it has been Joint Commission International (JCI) accredited since 2006 and is the first hospital system in the region to achieve institutional accreditation from the Accreditation Council for Graduate Medical Education-International (ACGME-I). Currently, there are three adult ICUs in Hamad General Hospital, namely, Medical ICU (MICU) with 22 beds, Surgical ICU (SICU) with 12 beds, and Trauma ICU (TICU) with 15 beds.

2.2.
Definitions. ABM was diagnosed on the basis of at least one of the following compatible clinical pictures with no other apparent cause: fever (38 ∘ C), headache, meningeal signs, cranial nerve signs, and impaired mental status, plus one of the following [7,8]: (1) Positive cerebrospinal fluid (CSF) culture (2) Positive CSF bacterial antigen test (with latex agglutination counterimmunoelectrophoresis) associated with pleocytosis mainly neutrophilic, defined as absolute WBC ≥ 100 cells/mm 3 , with a decreased glucose level ≤ 40 mg/dL and an increased protein concentration ≥ 60 mg/dL.
ABM was considered nosocomial if the diagnosis was made after more than 48 hours of hospitalization or within a short period of time (i.e., usually within one month after discharge from the hospital where the patient had received an invasive procedure, especially a neurosurgical procedure) [9]. On the other hand, ABM was considered as communityacquired if the diagnosis was made within the first 48 hours of hospitalization and the patient was not hospitalized in the preceding month [10]. Empirical antimicrobial therapy was deemed to be inappropriate if the antibiotics were administered more than 24 hours after CSF collection and/or when the dosage, route, and duration of treatment were not in accordance with hospital guidelines [11]. Hypotension was defined as blood pressure < 90/60 mmHg. Multidrugresistant organisms are defined as microorganisms, predominantly bacteria, that are resistant to one or more classes of antimicrobial agents [12]. Viral, fungal, mycobacterial, polymicrobial, and drug induced meningitis were excluded. ABM episodes with the same organism were included only once. Coagulase-negative staphylococci and viridans streptococci are considered as causative agents if CSF showed pleocytosis mainly neutrophilic, defined as absolute WBC ≥ 100 cells/mm 3 , or a decreased glucose level ≤ 40 mg/dL or an increased protein concentration ≥ 60 mg/dL. The primary outcome was inhospital mortality which included all causes of death during admission.

Isolation, Identification, and Antimicrobial Susceptibility Test of Microorganisms.
Identification of isolates was based on colony morphology, Gram stain, oxidase, catalase, VITEK 2 Compact (bioMérieux, Durham, USA), and Phoenix (Becton Dickinson, NJ, USA). The antimicrobial minimal inhibitory concentrations (MICs) for the isolates were determined by using Phoenix (Becton Dickinson, NJ, USA) for GNB and staphylococci and enterococci (among Gram-positive cocci). For fastidious bacteria, susceptibility was determined with a gradient strip method ( -test strips, bioMérieux, Marcy-l'Étoile, France). The breakpoint interpretation was determined according to the recommendations of the Clinical Laboratory Standards Institute (CLSI) [13].

Source of Data and Data
Collection. Cases were identified via hospital's discharge records, infection control records, and cerebrospinal fluid records maintained by the microbiology unit. These records were reviewed carefully by two investigators, in order not to miss any case. Records of all patients with bacterial meningitis were reviewed retrospectively to retrieve data on patients' demography, sign-symptoms, underlying medical conditions, investigations, names of microorganisms and their drug susceptibility, name and duration of therapy offered, appropriateness of therapy, and outcome.

Statistical Analysis.
Quantitative variables were expressed as mean ± SD. Univariate logistic regression was performed to determine the probable predictors of inhospital mortality. All potential risk factors at ≤0.1 level in the univariate analysis were entered in the multiple logistic regression to identify the independent predictors of mortality at < 0.05. The data were analyzed with SPSS software (v 17; IBM Corp., Armonk, NY, USA).

Trends of Antimicrobial Susceptibility.
Details of antimicrobial susceptibility are shown in Tables 5 and 6. Among the Gram-positive cases, 3 (18.6%) episodes of Streptococcus pneumoniae were resistant to ceftriaxone, while out of all coagulase-negative staphylococci isolates, 19 (95%) were methicillin-resistant. Among the Gram-negative cases, 100% of Chryseobacterium species were resistant to meropenem and colistin, while 63.3% of Acinetobacter species showed resistance to meropenem but none for colistin. All Pseudomonas spp. were sensitive to piperacillin-tazobactam and meropenem. Among Klebsiella isolates, 2 (16.6%) were extended spectrum beta-lactamase (ESBL) producers, but all    Table 7). Only the presence of underlying diseases, hypotension, and inappropriate treatment were found to be independent predictors of mortality by multivariate logistic regression analysis (see Table 8).

Discussion
Acute bacterial meningitis is a serious disease which necessitates early diagnosis and aggressive therapy to improve prognosis. Regional information regarding demographic data of patients, associated underlying conditions, etiology, and antimicrobial susceptibility is essential for correct and timely management of this disorder. Our study was the first to attempt to determine the clinical picture and the spectrum of pathogens of bacterial meningitis in patients of all ages in Qatar.
This retrospective series revealed some observations that deserve attention: firstly, in contrast with the previous study [6], the trend was seen to decrease from 2009 to 2013. Among the total 117 episodes, 43 (36.7%) were reported in the year 2009, which decreased to 18 (15.4%) in 2013. Furthermore, the disease in our series predominantly affected adults rather than infants and young children. This picture is similar to what was found in west countries and it may be attributed  Microorganisms  TNP  pen  amp  oxc  eryt  clind  amclv  cotr  cfr  van line teic  Abiotrophia spp.  1  0  0  NT  0  0  0  0  NT  0  NT NT  Enterococcus faecalis  7  NT  0  NT  NT  NT  NT  NT  NT  0  0  0  Enterococcus gallinarum  1  NT  0  NT  NT  NT  NT  NT      Thirdly, compared with the previous studies [4,6], changes of common causative pathogens of ABM had been noted in our series. Coagulase-negative staphylococci species were the most common causative agents followed by Streptococcus pneumoniae. This can be explained by the expansion of neurosurgical services in our hospital with a consequent increase in the number of patients with postneurosurgical state. Similarly, reports from Taiwan [15][16][17][18] showed that there has been an increasing incidence of staphylococcal infection in ABM patients. However, in agreement with many reports worldwide [2,7,[20][21][22][23], Streptococcus pneumoniae remain the common causative agent for community-acquired infection in our study.
Fourthly, drug resistance pattern showed that 95% of the implicated coagulase-negative staphylococci species were oxacillin-resistant and 63.3% of the implicated Acinetobacter species were meropenem-resistant. Both infections were predominantly nosocomial, which raised doubt regarding the infection control program in our hospital. Moreover, these findings result in therapeutic challenge in the choice of empiric antibiotics in the initial management of ABM. These findings are consistent with reports coming from Taiwan recently [9,18,19]. Fortunately, so far, we have not encountered vancomycin-resistant coagulase-negative staphylococci strains or colistin-resistant Acinetobacter strains.
Finally, in an attempt to identify independent predictors of mortality in patients with ABM, many studies had been conducted. The concluded prognostic factors among these studies were diverse [9,10,18,19,22,23]. Our study revealed many probable prognostic factors; however, only the presence of underlying diseases, hypotension, and inappropriate treatment were found to be independent predictors of mortality by multivariate logistic regression analysis.
This hospital-based study has the following limitations. First, the study was retrospective rather than prospective, and this design did not allow us to obtain additional details such as severity of the disease and long-term follow-up to evaluate the long-term sequelae of meningitis in our patients. Second, it was performed at a single hospital; the results may not be applicable to other hospitals. Third, we included patients who had a positive CSF culture or positive CSF bacterial antigen test.
Despite these limitations, we believe that our study remains the largest to date to provide comprehensive information on the epidemiology of ABM in Qatar.
In conclusion, our study revealed that there is a change in the predominantly affected age group and common causative agents of ABM. Coagulase-negative staphylococci species are the common causative agent in Qatar with majority of infections occurring nosocomially. More than 90% of all implicated coagulase-negative staphylococci strains were oxacillin-resistant. Thus, improving our infection control programs in addition to enhancing antimicrobial stewardship is essential to overcome this problem.