NSAIDs, pneumonia and empyema: an ongoing confrontation
Over the past twenty years, the worldwide increase in cases of pleural empyema and complicated pneumonia by Streptococcus pneumoniae in children has attracted much epidemiological and clinical attention.1 In fact, particularly invasive and/or pathogenic serotypes have emerged – especially serotypes 1, 19A, 3, 14 and 7F – absent in the commonly used heptavalent vaccine (PCV7) and then included in the 13-valent vaccine, which has been improved with even more serotypes in the 23-valent. The causes behind this increase are not completely clear yet. The hypothesis that the PCV7 might have caused a progressive positive selection of previously epidemiologically marginal strains has been confuted by the increase of serotype 1 in Spain, observed well before the introduction of PCV7.2 It has been also suggested that the increased use of NSAIDs – ibuprofen in particular – in case of respiratory infections or pneumoniae could have favoured a higher bacterial invasiveness and, then, the onset of empyema.3-7 The possibility that ibuprofen could affect negatively the course of some bacterial infections had already been studied in children, especially in relation to the risk of necrotising fasciitis caused by Group A Streptococcus secondary to varicella (relative risk 4.9) or zoster virus (relative risk 1.6) or, at least, to invasive infections with combined treatment of ibuprofen and acetaminophen.
One of the first report, back in 2002, started from the evidence of a significant increase of empyema incidence in the children hospitalised for pneumonia and tried to identify some risk factors, not exclusively microbiological.3 It found a significant correlation with the onset of empyema in case of: history of varicella (odds ratio 14.0), ≥7 days of fever before the hospitalisation (odds ratio 6.4), age ≥3 years (odds ratio 4.0), chest pain (odds ratio 2.0) and home administering of ceftriaxone (odds ratio 3.3) or ibuprofen (odds ratio 4.0). The reasonable interpretation of these data is that low dosage ceftriaxone in single intramuscular administration or ibuprofen for the treatment of fever had to a certain measure masked the evolution of pneumonia, delaying the diagnosis and the treatment of empyema. To similar conclusions came a study on adult subjects: who had taken NSAIDs at home had a higher incidence of empyema and lung abscess (37.5% in respect to 7%, odds ratio 8.1) and, if an antibiotic had not been used at the same time, a higher risk (odds ratio 3.8) of invasive disease (bacteraemia or empyema).6 Also in this case the authors hypothesise a NSAIDs interference with the infection in its initial stages (modification of neutrophil and alveolar macrophages functionality, alteration of inflammatory processes linked to arachidonic acid derivatives) and a “make-up” effect on the wellbeing state that would end up delaying the correct overview of the complications. Further observations, still in paediatric age, confirm a certain association between the use of ibuprofen at home for controlling fever and discomfort and further evolutions of complicated pneumonia.4,5 When the responsible pathogen had been isolated, in 50-70% of cases it was Streptococcus pneumoniae and, less frequently, other bacteria among which Group A streptococci, Staphylococcus aureus, fusobacteria, virus or mixed infections. Lastly, an increased risk of empyema (odds ratio 2.79) has been reported also in a recent multi-centre control-case study in children treated with NSAIDs in course of acute respiratory virus infection.10
Evidences of a NSAIDs causal role in complicated pneumonia are not strong, but deserve attention because of the experimental and pharmacological rationale that attributes to NSAIDs a depressive effect upon some important neutrophils functions (chemotaxis, adhesion, aggregation, degranulation) and the inhibition of prostaglandin synthesis.8 All this could lay the foundation for a relative state of immune depression that may favour pathogens invasiveness. On the contrary, fever enhances immune defences and reduces virus and bacteria replication and many serotypes of Streptococcus pneumoniae are thermolabile and do not survive at 40-41°C. Role and treatment of fever in the child are up to date object of interesting reflections still.11 The possible bias of observations are nevertheless numerous in all the cited studies and regard NSAIDs administering timing and doses, the co-use at home of different types and dosages of antibiotics, the timing and the diagnostic criteria used to detect complications and, lastly, the type (or serotype) of identified pathogen. The clear correlation between the increase in sales of paediatric formulation of ibuprofen and the reported cases of complicated pneumonia in France (see figure) is quite impressive even though not necessarily causal.4
and Enrico Valletta, Paediatrics Unit, Hospital GB Morgagni -L. Pierantoni, AUSL Romagna, Forlì
- Eur J Clin Microbiol Infect Dis 2014;33:879-910 CDI
- Clin Microbiol Infect 2011;17:1441-4 CDI
- Clin Infect Dis 2002;34:434-40 CDI NS
- Acta Paediatr 2010;99:861-6 CDI
- Pediatr Pulmonol 2015;50:721-6 CDI
- Chest 2011;139:387-94 CDI
- BMJ Case Rep 2013;DOI:10.1136/bcr-2013-200544
- Br J Clin Pharmacol 2007;65:203-9
- Pediatrics 2001;107:1108-15 CDI NS
- J Pediatr 2016;175:47-53
- Arch Dis Child 2015;100:818-20 CDI