BACTERIAL-MENINGITIS IN CIRRHOSIS - REVIEW OF 16 CASES

Citation
A. Pauwels et al., BACTERIAL-MENINGITIS IN CIRRHOSIS - REVIEW OF 16 CASES, Journal of hepatology, 27(5), 1997, pp. 830-834
Citations number
34
Categorie Soggetti
Gastroenterology & Hepatology
Journal title
ISSN journal
01688278
Volume
27
Issue
5
Year of publication
1997
Pages
830 - 834
Database
ISI
SICI code
0168-8278(1997)27:5<830:BIC-RO>2.0.ZU;2-7
Abstract
Background: Although bacterial infections are frequent in patients wit h liver cirrhosis, only isolated cases of bacterial meningitis have be en reported. Methods: We have reviewed a series of 16 cases of bacteri al meningitis in patients with cirrhosis, diagnosed in a single hospit al over a 30-year period. Results: Thirteen patients had alcoholic cir rhosis. On presentation, all patients had fever and most of them had a n abnormal mental status (coma in 11 cases), but neck stiffness was no t present or was delayed for more than 24 h in seven (43.7%) patients. The cerebrospinal fluid white cell count was always elevated, higher than 1000/mu l in ten cases. The cerebrospinal fluid culture was posit ive in 14 (87.5%) patients. Gram-negative bacilli (mainly E. coli) and L. monocytogenes were the most frequent pathogens, accounting for nin e cases. In contrast, S. pneumoniae and N. meningitidis were found in only four cases. Concurrent bacteremia was present in 12 (75%) cases. Ten patients (62.5%) died. Death was meningitis-related in seven patie nts and due to decompensated liver cirrhosis after clinical recovery f rom meningitis in tile three other patients, Child-Pugh class C was as sociated with a higher mortality rate (80%, versus 33% for Child-Pugh class A-B), although the difference did not reach statistical signific ance. Conclusions: Bacterial meningitis should be suspected in every p atient with cirrhosis presenting with a febrile coma, If lumbar punctu re must be delayed, or if no causative agent can be identified on cere brospinal Gram stain despite elevated cerebrospinal fluid white cell c ount, empirical antimicrobial therapy should be started straightaway w ith ampicillin plus a third-generation cephalosporin in sufficient dos es.