The epidemiology of peritonsillar abscess disease in Northern Ireland

1. To describe the epidemiology of peritonsillar abscess disease in Northern Ireland. 2. To describe the impact of the nature of microbiological sampling on culture results.

Retrospective review of cases of peritonsillar abscess identified by diagnostic coding in three centres in Northern Ireland between August 2001 and July 2002.

One hundred and twenty eight cases of confirmed peritonsillar abscess were treated as inpatients accounting for 1 in 10,000/year of the population in the hospitals' catchment area. The mean age was 26.4 (range 9-78) years. Sixty-nine patients were male; the mean length of hospital stay was 3 days. Culture yield was greatest from needle aspirates; throat swabs and blood cultures were typically unhelpful. Beta haemolytic streptococci were the most common isolates. Resistance among Group A haemolytic streptococci to macrolide antibiotics was present in 26% of isolates. Heterophile antibody testing was routine and revealed that Epstein-Barr virus infectious mononucleosis had a prevalence of 1.8% in this group of patients.

In this population, although there were many similarities with the clinical features reported by other investigators, there were also several differences. Notably, we found a comparatively low proportion of anaerobic infections. Although cultures results did not influence individual patient treatment, reviewing data derived from populations is valuable for guiding empirical antibiotic therapy.

The epidemiology of peritonsillar abscess disease is not well described. Other investigators describing the bacteriology give differing results, presumably reflecting different countries and techniques. The need for routine microbiological testing has been questioned in several reports, but the benefit of routine culture and sensitivity data for planning empirical antibiotic treatment has not been explored. We present a retrospective review of 128 cases of confirmed peritonsillar abscess in Northern Ireland from August 2001 to July 2002. The annual population incidence in this region is 1 in 10,000/year. In this group concurrent Epstein-Barr virus infection was found to have a prevalence of 1.8%. The yield of positive cultures from routine microbiological sampling was low in comparison to that reported in detailed prospective studies, from various countries. Needle aspiration was the best technique for obtaining pus for culture; throat swabs and blood cultures were unhelpful. Culture results did not affect individual patient treatment, but may have a potentially useful role for selecting empirical therapy.