Premium
Unsuspected bacterial infections in febrile convulsions
Author(s) -
McLntyre Peter B.,
Cray Susan V.,
Vance John C.
Publication year - 1990
Publication title -
medical journal of australia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.904
H-Index - 131
eISSN - 1326-5377
pISSN - 0025-729X
DOI - 10.5694/j.1326-5377.1990.tb125143.x
Subject(s) - leukocytosis , medicine , urinary system , meningitis , lumbar puncture , bacteremia , prospective cohort study , white blood cell , pediatrics , surgery , cerebrospinal fluid , antibiotics , microbiology and biotechnology , biology
in a 12‐month prospective study in 1984, blood and urinary cultures were obtained as a routine from 307 children who presented with fever and convulsions to the Mater Misericordiae Children's Hospital, Brisbane, and the results were compared with data from 1981–1983 when cultures were not taken as a routine. In the prospective study, bacteraemia was found in 12 (4.3%) of 282 patients but was not suspected clinically in half of these; urinary‐tract infection was found in seven (2.6%) of 272 patients and in six of these it was not suspected clinically. All 12 patients with unsuspected bacteraemia or urinary‐tract infection had persistent fever; of these, nine patients suffered simple convulsions and all cases of urinary‐tract infection occurred in female patients. Bacteraemia was significantly more common in patients of less than two years of age, in children who were selected for lumbar puncture and in the study period compared with the retrospective period, 1981–1983. Leukocytosis (white‐cell count, more than 15.0 times 10 9 /L) was a sensitive (75%) diagnostic aid but was poorly specific (59%) for bacteraemia. Bacterial meningitis was not diagnosed initially in four of the nine cases which occurred among children who presented with fever and convulsions between 1981 and 1984; in all four children, the cerebrospinal fluid appeared normal at hospital admission. We conclude that bacteraemia and urinary‐tract infections are detected more frequently in children who are admitted to hospital with febrile convulsions when cultures are performed as a routine. In the at‐risk group (children of less than two years of age), the prevalence of urinary‐tract infection is increased in female patients and the prevalence of bacteraemia is increased in those patients who are selected for lumbar puncture. The use of leukocytosis as a criterion to determine the need for blood cultures improves the diagnostic yield but would result in increased costs and additional venepuncture. Bacterial meningitis was rare in our case series and the performance of a lumbar puncture as a routine at admission to hospital would not have led to its earlier diagnosis.