Prognostic factors in pediatric pneumococcal meningitis patients in mainland China: a retrospective multicenter study.

Abstract

Background: Prognosis of pneumococcal meningitis (PM) remains very poor, especially in less developed countries. Currently, few multi-centric studies on pediatric PM have been reported in mainland China.Objectives: This study aimed to explore the correlation of clinical and laboratory findings with complications and prognosis in pediatric PM.Methods: The pediatric PM patients were retrospectively recruited from ten pediatric tertiary hospitals across China between January 2013 and June 2018. Clinical, biochemical, and microbiological data and follow-up information were collected. Predictive factors for complications and prognostic factors for overall survival (OS) and sequelae-free survival (SFS) were analyzed.Results: A total of 132 pediatric PM patients were included. Seventy-one patients had complications, 25 patients died, and 39 patients had neurological sequelae. Multivariate logistic regression suggested that age less than 28 months (adjusted OR=2.654, 95%CI=1.067–6.600, P=0.036) and lower white blood cells in blood (aOR=3.169, 95%CI=1.395–7.202, P=0.006) were associated with high risk of complications. Multivariate Cox’s proportional hazard regression suggested that age less than 28 months (aHR=6.479, 95%CI=1.153–36.404, P=0.034), coma (aHR=9.808, 95%CI=2.802–34.323, P=0.000), and non-adjuvant steroid therapy (aHR=4.768 95%CI=1.946–11.678, P=0.001) were independent prognostic factors for poor OS; coma (aHR=5.841, 95%CI=2.652–12.864, P=0.000), septic shock on admission (aHR=2.949, 95%CI=1.049–8.290, P=0.040), and lower glucose level in cerebrospinal fluid (CSF) (aHR=2.523, 95%CI=1.336–4.765, P=0.004) were independent prognostic factors for poor SFS.Conclusion: Age, coma, and adjuvant steroid therapy were independent factors for OS, while coma, septic shock on admission, and lower glucose level in CSF were independent factors for SFS in pediatric PM patients. These factors might be used to identify PM patients with poor prognosis and guide individual treatment.