Introduction
Acute epiglottitis is a potentially life-threatening infection with septic characteristics. It was previously considered to be a childhood disease. A typical feature of the disease was a drooling child sitting upright and leaning forward to make breathing easier. However, during the 1980s cases in adults began to be reported.Reference MayoSmith, Hirsch, Wodzinski and Schiffman1 After the introduction of general Haemophilus influenzae type b (Hib) vaccination in children, the disease became rare, and now affects adults almost exclusively.Reference Guldfred, Lyhne and Becker2 In adults, examination can be performed either indirectly with a larynx mirror or (preferably) directly with a flexible nasolaryngoscope. The disease is treated with intravenous antibiotics (cefotaxime) and by securing the airway in an intensive care unit. Airway management often requires endotracheal intubation. Before general Hib vaccination was introduced in Sweden in 1993, the incidence of epiglottitis was 10–14 cases/100 000/year for children (≤14 years), 1.8–2.3 cases/100 000/year for adults and 3.2–4.5 cases/100 000/year for all ages.Reference Hugosson, Olcén and Ekedahl3, Reference Trollfors, Nylén and Strangert4 In the general vaccination programme, children are vaccinated at three, five and 12 months of age.
The aims of this study were (1) to investigate the incidence and age distribution of acute epiglottitis in a well defined population, following the introduction of general childhood Hib vaccination, and (2) to study changes in the bacterial epidemiology of the disease. For comparison, a corresponding study in the same population has been published, which was conducted before the introduction of Hib vaccination.Reference Hugosson, Olcén and Ekedahl3 The age-specific incidence found by the pre-vaccination study is shown in Table I. In a preliminary study of this population conducted in the first few years of the general Hib vaccination programme, we found a decreased overall incidence of acute epiglottitis, along with increased aetiological importance of Streptococcus pneumoniae.Reference Isakson and Hugosson5 Consequently, an objective of the present study was to analyse the S pneumoniae serotypes involved in acute epiglottitis.
Materials and methods
We reviewed the medical records of patients from Örebro County, Sweden, who were diagnosed with acute epiglottitis (International Classification of Disease code J05.1) between 1993 and 2008 at Örebro University Hospital. In Örebro county, all patients with acute epiglottitis were referred to and treated at the Örebro University Hospital; we therefore assumed that all patients from the county who were treated for acute epiglottitis during this period were identified.
We included in the study patients who had experienced difficulties in swallowing or breathing for less than four days, who had a fever of more than 37.9°C, and who had signs of inflammation of the epiglottis as verified by direct or indirect laryngoscopy. Patients were excluded if they had an inflammatory reaction of the epiglottis which was judged to be part of another disease, for example an infection originating from the tonsils. These criteria were identical to those applied in the study of the same population conducted before general Hib vaccination.Reference Hugosson, Olcén and Ekedahl3
During the study period, 60 patients were diagnosed with acute epiglottitis as defined above. Seventeen of these patients were excluded as they did not meet the inclusion criteria. Patients were excluded due to: failing to meet the temperature criterion (i.e. more than 37.9°C; 13 patients); having an infection originating from the tonsils (two patients); and falling ill more than four days before admission to hospital (two patients).
The S pneumoniae strains identified in patients with positive blood cultures were serotyped at the Swedish National Institute for Disease Control, using gel precipitation with a panel of antisera.
Demographic information about Örebro county was obtained from the Statistics, Sweden database.6
Results
Forty-three patients (15 women, 28 men) were included, 91 per cent of whom (39/43) were adults. The age-specific incidence of acute epiglottitis during the first 16 years of the Hib vaccination era is shown in Table I. The age distribution is shown in Figure 1. Four children suffered acute epiglottitis between 1993 and 1995; none had been vaccinated against Haemophilus influenzae type b (Hib). From 1996 onwards, there were no cases of childhood acute epiglottitis.
Age distribution of patients with acute epiglottitis in Örebro county, 1993–2008 (i.e. the first 16 years of general Haemophilus influenzae type b vaccination).
Blood cultures were taken from 41 patients. These were positive in 44 per cent (18/41) of cases. The bacterial species found in the positive blood cultures are shown in Table II. Streptococcus pneumoniae was found in 24 per cent (10/41) of cases and Hib in 15 per cent (six of 41). In seven of the 18 patients with a positive blood culture, the same bacteria could be identified in a nasopharyngeal culture; in eight cases, the nasopharyngeal culture was negative. In one case, a different bacterial species was found in the nasopharyngeal culture compared with the blood culture. In two cases, no nasopharyngeal culture specimens were taken. No Hib strain produced β-lactamase, and all pneumococcal strains were sensitive to penicillin G and penicillin V. Nine of the 10 pneumococcal strains could be serotyped; these comprised serotypes one (one strain), four (three strains), seven (two strains), 12 (two strains) and 38 (one strain).
Bacterial aetiology* of acute epiglottitis, örebro county, 1993–2008
* Confirmed by blood culture. †Negative blood culture.
Table III presents the agent-specific incidence of acute epiglottitis in adults, before and after the introduction of general Hib vaccination.
Complications of acute epiglottitis occurred in three adult patients (7 per cent). One patient developed an abscess, while another developed an epiglottal retention cyst. The third patient had serious inspiratory distress during attempted extubation, requiring tracheotomy. No deaths occurred.
Discussion
Streptococcus pneumoniae was the major cause of acute epiglottitis in this study, in accordance with previous reports from the Haemophilus influenzae type b (Hib) vaccination era.Reference Isakson and Hugosson5, Reference Wood, Menzies and McIntyre7 In the present study population, S pneumoniae was the causative agent in 24 per cent of cases; between 1975 and 1992, this figure was 3 per cent. Therefore, there has been a 2.8-fold increase in the incidence of S pneumoniae epiglottitis in the adult population, comparing the two studies.Reference Hugosson, Olcén and Ekedahl3 Of the S pneumoniae strains identified, eight of the nine tested serotypes are represented in the 23-valent polysaccharide vaccine used in adults for the prevention of invasive pneumococcal infection.Reference Bogaert, Hermans, Adrian, Rümke and de Groot8, Reference Moberley, Holden, Andrews and Tatham9 Only one strain (serotype 38) is not represented in this vaccine.
The change in bacterial epidemiology does not change the antibiotic policy for acute epiglottitis treatment. The antibiotic used should have coverage for both β-lactamase producing Hib as well as S pneumoniae with reduced sensitivity for penicillin G. Cefotaxime is therefore a good initial choice. Subsequently, the antibiotic selection should be guided by the blood culture results and antibiotic resistance pattern.
The incidence of acute epiglottitis of unknown aetiology (i.e. with a negative blood culture) was reduced by 50 per cent in this study, compared with pre-vaccination findings.Reference Hugosson, Olcén and Ekedahl3 This can be explained if a major proportion of infections with a negative blood culture are in fact caused by Hib. One should remember that a single dose of oral antibiotics taken before blood sampling can result in a negative blood culture. This demonstrates the difficulty of identifying the causative micro-organism of septic disease from a small blood sample culture. In the present study, the poor correlation between blood and nasopharyngeal culture results indicates that the bacterial cause of epiglottitis may not be verified by pharyngeal culture samples. Detection of bacterial antigens or DNA in blood or urine may develop as a complementary test, in the quest to improve the diagnostic success of acute epiglottitis bacteriology.Reference Trollfors, Nylén, Carenfelt, Fogle-Hansson, Freijd and Geterud10
Today, acute epiglottitis is a disease which affects adults almost exclusively. In the present study, only 9 per cent of affected patients (four of 43) were children; before the introduction of general Hib vaccination for infants, this figure was 58 per cent (126/219).Reference Hugosson, Olcén and Ekedahl3 In our population, no child suffered acute epiglottitis after 1996. Even if the proportion of adult acute epiglottitis cases has increased, compared with childhood cases, the overall incidence of the disease in the adult population has decreased to less than half of that of the pre-vaccination era.Reference Hugosson, Olcén and Ekedahl3
These results confirm the beneficial effect of general childhood Hib vaccination on the occurrence of acute epiglottitis in both children and adults. In our study, no patient had received Hib vaccination. However, it is important to recognise that Hib-vaccine does not have complete coverage, and that case-reports of acute epiglottitis and other invasive Hib-infections in vaccinated children have been reported.Reference Garpenholt, Hugosson, Fredlund, Giesecke and Olcén11, Reference McEwan, Giridharan, Clarke and Shears12
• General Haemophilus influenzae type b vaccination has dramatically reduced the incidence of acute epiglottitis in both children and adults
• However, in adults the agent-specific incidence of acute epiglottitis caused by Streptococcus pneumoniae has more than doubled
• The possibility of preventing adult acute epiglottitis by vaccination mandates further study of the bacterial epidemiology of all invasive S pneumoniae infections in adults
In future, the incidence of Hib acute epiglottitis will probably continue to decrease, as the greater part of the population becomes vaccinated against Hib and herd immunity increases.Reference Mühlemann, Alexander, Pepe, Weiss and Schopfer13 However, there is reason to believe that the number of cases of S pneumoniae acute epiglottitis will increase in the biological niche created by the absence of Hib. In the studied adult population, the number of cases of S pneumoniae acute epiglottitis had already more than doubled, compared with the pre-vaccination era. The 23-valent polysaccharide vaccine is currently offered to immunocompetent adults at increased risk of invasive S pneumoniae infection, and also to splenectomised patients. Risk factors for invasive pneumococcal pneumonia are considered to comprise: an age of more than 65 years, cardiac failure, chronic lung disease, chronic liver disease, unstable diabetes mellitus and elderly patients cared for in institutions. The evidence is poor for an effect of the vaccine on immuno-incompetent patients, but it may still be recommended in this group.Reference Bogaert, Hermans, Adrian, Rümke and de Groot8, Reference Moberley, Holden, Andrews and Tatham9
Conclusion
Acute epiglottitis is a rare disease that affects adults almost exclusively. Streptococcus pneumoniae is the major causative bacterial agent. In the current study, bacterial serotyping indicated that a large proportion of serious infections could be prevented by use of the 23-valent polysaccharide vaccine. These findings show the importance of continued epidemiological analysis, including serotype assessment, not only of pneumonia but also of epiglottitis and other diagnoses involving S pneumoniae bacteraemia in the adult population. The results of such analysis are important to consider in future dicussions about appropriate vaccination programmes, to prevent invasive S pneumoniae infection in adults.
