• N Wood, R Menzies, and P McIntyre.
• National Centre for Immunization Research and Surveillance of Vaccine Preventable Diseases, Westmead, New South Wales, Australia. nicholw3@chw.edu.au
• Intern Med J. 2005 Sep 1;35(9):530-5.

BackgroundAcute epiglottitis due to infection with Haemophilus influenzae type b (Hib) is much less common in children following the introduction of Hib vaccination; however, adult epiglottitis cases have not decreased. In addition, epiglottitis hospitalizations are consistently more numerous than notifications and the reason for this is not clear.AimsTo more accurately describe the clinical, aetiological and epidemiological features of epiglottitis and to ascertain the accuracy of hospitalization data in an era of widespread Hib vaccination.MethodsMedical records in 11 public hospitals in three area health services in New South Wales with a principal or stay diagnosis (International Classification of Diseases (ICD)-9-CM or ICD-10-AM code) of acute epiglottitis between July 1990 and June 1992 (prior to Hib vaccination = pre-vaccine era) and July 1998 and June 2000 (widespread Hib vaccination = vaccine era) were reviewed. Case definitions of epiglottitis were applied.ResultsOne hundred and forty-two records were identified (114 pre-vaccine era and 28 vaccine era). Incorrect coding was more common in vaccine era records (32 vs 7%). Of correctly coded records, adults over 20 years old comprised the majority in the vaccine era (84 vs 17%). Hib bacteraemia was identified in 62% of cases in the pre-vaccine era compared to no cases in the vaccine era, despite equivalent blood cultures being taken between the two eras (84 vs 74%). Streptococcus pneumoniae was the only other organism isolated. Three deaths were recorded (1 child, 2 adults), all in the pre-vaccine era.ConclusionsAcute epiglottitis hospitalizations in the current Hib vaccine era are predominantly in adults, and rarely are Hib or other causative organisms identified, although microbiological data are often incomplete. The discrepancy between hospitalization and notification data appears to be due to misclassification of hospitalization records.

28 keywords...

hide…