Should all patients with influenza be prescribed antibiotic prophylaxis?
Date First Published:
July 20, 2009
Last Updated:
July 20, 2009
Report by:
Craig Haddock, Medical student (Manchester University)
Search checked by:
Craig Haddock, Manchester University
Three-Part Question:
In [healthy adults with influenza A or an influenza-like illness presenting in the emergency department] is the use of [antibiotic prophylaxis] of [proven benefit]?
Clinical Scenario:
A 30 year old man presents in the emergency department with an influenza-like illness. You wonder whether or not antibiotic prophylaxis would be of overall benefit to them with regard to duration of illness and incidence of bacterial complications.
Search Strategy:
MEDLINE (1950 to June week 2 2009) and EMBASE, via OVID interface
({[exp Anti-Bacterial Agents/] AND [exp Antibiotic Prophylaxis]} OR {[exp Anti-Bacterial Agents/] AND [prophylaxis.mp.]}) AND {[exp Influenza, Human/] OR [exp Influenza A virus, H1N1 Subtype/]} LIMIT to [English language and human]
({[exp Anti-Bacterial Agents/] AND [exp Antibiotic Prophylaxis]} OR {[exp Anti-Bacterial Agents/] AND [prophylaxis.mp.]}) AND {[exp Influenza, Human/] OR [exp Influenza A virus, H1N1 Subtype/]} LIMIT to [English language and human]
Search Details:
The MEDLINE search returned 32 papers, 31 of which were irrelevant to the proposed question and EMBASE showed no other papers. The remaining paper was a review which was relevant to the question. The references of this paper were then searched which revealed two papers which were deemed relevant to the question posed. There was a low level of study type that would be accepted for this question as the level of evidence available was low. It should be noted that some papers may have been missed due to inability to translate from French, although some did publish English abstracts.
Relevant Paper(s):
| Study Title | Patient Group | Study type (level of evidence) | Outcomes | Key results | Study Weaknesses |
|---|---|---|---|---|---|
| Using antibiotics in case of influenza. Chidiac, et al. 2006 France | Patients of all ages with bacterial complications of influenza. | Using antibiotics in case of influenza. | Frequency of occurrence of bronchitis and pneumonia in influenza infected population (%) | Bronchitis – 10 to 20% global population: Pneumonia – 1.8% global population | |
| Microbiological diagnosis of pneumonia in lethal cases of H1N1, H3N2 and influenza B (n=22) vs Asian influenza (n=536) (%) | S. aureus – 45% vs 69%, Mixed infection – 18% vs 23%, S. pneumoniae – 5% vs 9%, Strep. A and B – 14% vs 1% | ||||
| Mortality of pneumonia with S.aureus vs pneumonia of different aetiology (%) | 47% vs 16% (p<0.001) | ||||
| Frequency of antibiotic prescription in complicated cases vs uncomplicated cases (%), in all ages | The frequency of prescription is significantly higher in the event of complication (79,5 vs 41,6%: p < 0,0001), and for the ages 65 years and above | ||||
| Antibiotic therapy, clinical features, and outcome of 36 adults presenting to hospital with proven influenza: do we follow guidelines? Jones, A. et al 1991 UK | 36 elderly patients with confirmed influenza | Retrospective survey | Anti-staphylococcal therapy | 77% survivors (n=22) vs 50% non-survivors (n=14) | Retrospective study Sample – small size and specific Primary focus as audit, and does not relate directly to question posed |
| Symptoms (cough, discoloured sputum, dyspnoea, chest pain) | 91% survivors vs 43% non-survivors (p<0.05), 59% survivors vs 29% non-survivors, 86% survivors vs 43% non-survivors, | ||||
| Confusion on examination | 9% survivors vs 43% non-survivors (p<0.05) | ||||
| Lymphopenia | 47% survivors vs 89% non-survivors (p<0.05) | ||||
| Plasma urea > 7mmol/l | 21% survivors vs 73% non-survivors (p<0.05) | ||||
| Efficacy of antibiotics against influenza-like illness in an influenza pandemic Maeda, S. et al 1999 Japan | 85 children with influenza-like illness. | RCT | Incidence of complications associated with influenza | No significant difference in duration of fever, incidence of acute otitis media (p=1.00), or diarrhoea associated illness (p=0.216), Incidence of pneumonia less in antibiotic group than placebo group (2.4% vs 16.3%, p=<0.05) | Small study Little information provided about methods, including randomisation Specific patient group may not be generalised. |
| Antibiotic treatment for influenza does not affect resolution of illness, secondary visits, or lost workdays. Carrat, F. et al 2004 Netherlands | 701 patients with influenza-like illness. | Prospective cohort study. | Time to alleviation of symptoms (days) | No difference with use of antibiotics (7 days vs 8 days, Med CI 95%, p=0.31) | Non randomised and not blinded Lack of statistically significant results |
| Rate of secondary physician visits | No difference with use of antibiotics (24% vs 27%, p=0.22) | ||||
| Number of work days lost | No difference with use of antibiotics (3.7 days vs 3.8 days, p=0.97) | ||||
| Time to return to normal activity (days) | No difference with use of antibiotics (7 days vs 7 days, p=0.16) |
Author Commentary:
Maeda et al. found that complication rates of pneumonia in children aged between four months and 11 years and four months are decreased with the use of antibiotics. There was no significant difference between the duration of fever and incidence of otitis media. However, the incidence of pneumonia in the antibiotic treated group was significantly less than the untreated group. There is a wide range of ages and the table showing the ages of children involved in more depth was missing, therefore it is difficult to interpret the results specifically for the younger (< 2 year old), higher risk children. Also the small study group makes generalisation of the results difficult. Jones et al suggests that the high S. aureus infection rate and associated high mortality should emphasise the need for anti-staphylococcal cover in epidemic situations. However, the report is a retrospective study of a small, highly specific sample size of higher risk patients and has more emphasis on auditing past epidemics rather than answering the posed question.
Carrat et al postulated that there was no difference in time to alleviation of symptoms, rate of secondary physician visits, number of work days lost or time to return to normal activity between the antibiotic and non-antibiotic groups. The study was non-randomised and not blinded and therefore is a low quality of evidence and also some results are not significant. However, this is the largest study data available for the use of antibiotics in influenza and does indicate that there is inappropriate over-prescription of these drugs. Chidiac et al also suggest that there is an inappropriate over-prescription of antibiotics in the absence of confirmed bacterial complications. 41.6% of patients without complications were prescribed antibiotics. In the context of an epidemic this has marked adverse consequences with regard to both antibacterial-resistant strains emerging and the unnecessary gross expenditure for the NHS.
Carrat et al postulated that there was no difference in time to alleviation of symptoms, rate of secondary physician visits, number of work days lost or time to return to normal activity between the antibiotic and non-antibiotic groups. The study was non-randomised and not blinded and therefore is a low quality of evidence and also some results are not significant. However, this is the largest study data available for the use of antibiotics in influenza and does indicate that there is inappropriate over-prescription of these drugs. Chidiac et al also suggest that there is an inappropriate over-prescription of antibiotics in the absence of confirmed bacterial complications. 41.6% of patients without complications were prescribed antibiotics. In the context of an epidemic this has marked adverse consequences with regard to both antibacterial-resistant strains emerging and the unnecessary gross expenditure for the NHS.
Bottom Line:
There is insufficient evidence to suggest that antibiotic prophylaxis is beneficial in any patient without proven bacterial complications, and instead should be reserved for use in confirmed cases.
References:
- Chidiac, et al.. Using antibiotics in case of influenza.
- Jones, A. et al. Antibiotic therapy, clinical features, and outcome of 36 adults presenting to hospital with proven influenza: do we follow guidelines?
- Maeda, S. et al. Efficacy of antibiotics against influenza-like illness in an influenza pandemic
- Carrat, F. et al. Antibiotic treatment for influenza does not affect resolution of illness, secondary visits, or lost workdays.