Epidemiology of bacterial meningitis in children admitted to Gondar University Hospital in the post pneumococcal vaccine era

Introduction Community acquired bacterial meningitis (CABM) is responsible for high mortality and disabling sequelae. Introduction of pneumococcal conjugate vaccine (PCV-10) and haemophilus influenzeatype b (Hib) has changed the epidemiological and clinical features of patients presenting with CABM as it is shown in different literatures over the last decade. The aim of this study was to assess the clinical and epidemiologic features and outcomes of CABM after the introduction of PCV-10 in Gondar University Hospital (GUH). Methods This is a retrospective study among children between 2 months and 14 years of age discharged from Gondar University Hospital. All patient records discharged with a diagnosis of meningitis at GUH were reviewed from September 2011 - September 2013. The data was collected using a structured questionnaire from the patient record charts and analysis was done using SPSS-20. Results 80 cases (1.6%) of CABM out of 4996 admissions were identified. There were 60 (75%) cases of CABM using WHO criteria of cerebrospinal fluid leukocytosis (CSF) > 100cells/mm3, or 10-100cells/mm3 with either hypoglycorrhea or increased protein; and 20 (25%) with culture confirmation. S. Pneumoniae was the most frequent pathogen identified in 14 (70%) children. The most common age group were infants 2-12 month old (n = 32, 40%). Children with adverse outcomes had shown a higher frequency of being older children (p = 0.045), loss of consciousness (p = 0.046), seizure at admission (p < 0.01), and a positive CSF culture (p = 0.03). Conclusion Introduction of PCV-10 has shown a decreased admission rate, mortality, and neurologic sequelae due to CABM.


Introduction
Community acquired bacterial meningitis is a life threatening infection of the leptomeninges often related with serious complication and sequelae. Low and middle-income countries account for 98% of the estimated 5.6 million disability-adjusted life years attributed to meningitis globally. In high-income countries, bacterial meningitis ranks among the top ten causes of death in children younger than 14 years of age [1]. Primary prevention of meningitis using vaccines is paramount, since death and long-term disabling sequelae are substantial in all settings especially those with least access to health care [2]. Bacterial meningitis accounts for approximately 6-8% of hospital admissions in Ethiopia with a case fatality rate of as high as 22-28 [3]. The most common pathogens being S. pneumonia, N.
meningitides and H.influenzea [4][5][6]. During the last several decades disease epidemiology and clinical features has changed dramatically in the countries that adapted the conjugate vaccines against H.
influenza type b and S. pneumoniae [7][8][9][10][11].  Age related clinical and laboratory features of bacterial meningitis are shown in Table 2 and Table 3. There is no specific sign or symptom which can predict culture positivity as it's shown on analysis of symptoms versus the culture positivity rate Table 4. The majority of patients (n = 50, 62.5%) were treated with ceftriaxone and the rest with a combination of penicillin and chloramphenicol (n = 30, 37.5%). Just more than half of the patients (52.5%) were treated for 10 days and 28(35%) of them were treated for 14 days.

Methods
Dexamethasone was given for 56(70%) patients before the antibiotics or at the same time. Most patients (n=68, 85%) were discharged improved and the rest (n = 12, 15%) were discharged with sequelae, died or disappeared. Overall case fatality was 7.5% (6/80) and all the deaths were due to S.pneumoniae which gives a case fatality rate for this organism of 42.8%. All of the children who died were above the age of 10 years (P < 0.05). Prolonged fever is found to be more prevalent among children under five years followed by infants and older than 10 years (p = 0.36). Seizure after 72 hrs of admission was a common complication among children older than 10 years old (p = 0.283).

Different factors contributed for poor outcome in this study as it is
shown on

Discussion
The hot, dry seasonal increase in admission and male predominance was consistent with other studies from Bangladesh [13] and BurkinaFaso [14]; but not similar to previous studies from Gondar [4] and Addis Abeba [3].  [17][18][19][20]. In the present study, this frequency was 5% for the complication and mortality rate was 7.5%, which is significantly lower than seen in previous studies.
In previous studies loss of consciousness, focal neurologic deficit, seizure at admission, delayed presentation and S. pneumoniae were found to be risk factors for poor outcome [19][20][21][22][23]. In the present study, the presence of seizure at admission, loss of consciousness, culture positivity and focal neurologic deficit were associated with poor outcome which was in agreement to these previous studies.
However, use of dexamethasone and delayed presentation was not found to be a predictor of poor outcome in this study. Antibiotics of choice were found to have a strong association with death in the present study. This may reflect an increase in possible resistant strains in the community but it needs a laboratory baser surveillance study for to make conclusion. While third generation cephalosporins such as ceftriaxone are the first line treatment for CABM in our setting, it may not be available due to stock outs, resulting in health care providers using older treatment recommendations from WHO such as penicillin and chloramphenicol. The study was retrospective in design and subject to the potential biases of retrospective reviews. However, we collected objective clinical and laboratory features that should be reliably recorded in the medical record. The medical records also had limited data about the study patient's PCV-10 immunization status. Although overall immunization was high in the study we cannot exclude the possibility that some of our study patients were either unimmunized or under immunized with PCV-10.
The other limitation was the subjectivity of discharge conditions has created some difficulties in finding the exact outcome, especially of neurologic outcomes. Another limitation was the paucity of data regarding antibiotics susceptibility patterns, since laboratory reports were incomplete. Finally, this study had only 20% culture positive specimens, though this would result in an overestimation of CABM.
The limited availability of culture reflects many resource-constrained setting, and we believe by applying WHO criteria for the definition of CABM gives us a reasonable estimate of the burden of CABM in our community.

Conclusion
Based on the results of this study, we can conclude that introduction

Acknowledgments
We would like to thank Department of Paediatrics and Child Health, University of Gondar.