Prevalence, aetiology and antimicrobial susceptibility of bacterial neonatal meningitis at Tikur anbessa specialized hospital, Addis ababa, Ethiopia

Abstract

Background: Meningitis is inflammation of meninges, which affects all age groups from the newborn to elderly and occurs more commonly during the first month of life. The highest burdens of bacterial meningitis occur in an area of sub- Saharan Africa. Meningitis has been a problem in Ethiopia for the past decade and despite all the management advances the condition has remained constant. Objective: To determine the prevalence, aetiology and antimicrobial susceptibility of bacterial neonatal meningitis at Tikur Anbessa Specialized Hospital. Design and Setting: Descriptive cross sectional study which was conducted within 3 month of period at newborn unit of Tikur Anbessa Specialized Hospital, Addis Ababa, Ethiopia Methods: Study subjects were neonates hospitalized at TASH due to neonatal sepsis with parental/guardian informed consent. Neonates with contraindication to lumbar puncture (LP) and failed LP were excluded. Case definition for sepsis was presence of one or more of the following signs: feeding problem, lethargy, abnormal cardiovascular, respiratory and neurological signs, temperature instability or skin change. Confirmed neonatal bacterial meningitis was defined as isolation of bacterial pathogen from the cerebrospinal fluid (CSF) by culture and/or visualization by gram stain and probable bacterial neonatal meningitis if a neonate had the specified clinical signs of meningitis without a confirmation with culture or gram stain. Neonates were enrolled consecutively until we attained the minimum sample size. Questionnaire was used to collect data on socio demographic characteristics and clinical features of the study subjects. Laboratory result pro-forma was used in collecting CSF analysis. Lumbar puncture was performed on one hundred and seven neonates with sepsis before they were started on antibiotics or before the change to cephalosporin. Microscopy and cell count, culture and antimicrobial susceptibility tests were performed. Results: We enrolled 115 neonates with suspected meningitis of whom 8 were excluded due to contraindications to LP or failed LP. Male to female ratio was 1.7:1, 71 (66.4%) were admitted into the hospital before or at the age of 7days, and 42 (39.3%) were born with a low birth weight. Median birth weight was 2750gm [Interquantile range (IQR) 2000-3300]; median postnatal age was 3 days (IQR 2-13) and median gestational age 37weeks (IQR 36-38). Feeding intolerance (76.6%), lethargy (49.5%) and abnormal respiratory signs (37.4%) were the most common clinical features observed. White cell count was high in 12(11.2%) of cerebrospinal fluid samples. Bacteria were isolated in Cerebrospinal of six neonates, of which two isolates were Streptococcus Pneumoniae, and the other 4 isolates were Escherichia coli, Pseudomonas Aeruginosa, Klebsella Pneumoniae and Acinetobacter. We diagnosed probable meningitis among 11/107 (10.2%) neonates and bacteriologic confirmed meningitis among 6/107 (5.6%), giving overall prevalence of meningitis of 15.8% [95%CI= 8.8% - 22.7%]. Of the 11 neonates with clinical suspected meningitis 6 (35%) had detectable bacteria in CSF. All isolated bacteria were resistant to ampicillin and gentamycin but were sensitive to ceftriaxone and cefotaxime. Conclusion: Overall prevalence of meningitis among neonates with sepsis hospitalized at Tikur Anbessa Specialized Hospital, Addis Ababa, Ethiopia was 15.8% [95%CI = 8.9% - 22.7%]. Bacteria were detectable in 35.3% of neonates with clinical meningitis, and from 5.6% of all neonates with sepsis. Isolated bacteria were predominantly gram-negative. All bacterial isolates were resistant to ampicillin and gentamycin, most were sensitive to third generation cephalosporins.