4. DISCUSSION
This systematic review summarises the current evidence base for the use of antimicrobials in tracheostomy dependent paediatric patients. All 12 studies identified were retrospective cohort studies. These highlight three core themes that should be the topic of future research: i) route of antimicrobial administration (systemic versus topical); ii) the role of cultures in guiding antimicrobial use; and iii) timing of antimicrobial use (reactive versus prophylactic). All of which would help clinicians optimise management for these children, reducing the need for escalation of care and improving antimicrobial stewardship. There is an urgent need for prospective, randomised controlled trial evidence to address these issues and guide the management of this vulnerable patient group.
Systemic antimicrobials are in routine use for tracheostomy associated infections, with empirical treatment often aiming to coverPseudomonas and Staphylococcus as common causative organisms. Although anti-pseudomonal antimicrobial usage has been associated with increased hospital admission duration both on an individual level29 and hospital level28, this likely reflects other clinical confounders such as their use in severe infections. The relationship between longer hospital admissions and hospital level prescribing rates may also represent poorer antimicrobial stewardship leading to increased antimicrobial resistance.
Inhaled antimicrobial use is also widespread in the management of tracheostomy associated infections; this is largely based upon theoretical benefits, such as higher tracheal antibiotic concentrations and reduced side effects. As well as extrapolation of evidence from the management of adult ventilator associated pneumonia40, cystic fibrosis41, and non-cystic fibrosis bronchiectasis42. Clearly, none of these populations represent children with tracheostomy associated infections well due to differences in both pathology and between children and adults, including physiology, anatomy, and risk factor exposure profiles (such as smoking). Although nebulised antimicrobials have been used in the treatment of respiratory illness for over 70 years43, further studies are needed to evaluate and compare alternate antimicrobial use in children with tracheostomies 30.
Prophylactic inhaled antimicrobials are often used in clinical practice; however, there is little international consensus on their use in children with tracheostomies. To the best of our knowledge the only guidelines endorsing their use in this patient cohort are by the Brazilian Society of Paediatrics, specifically in the post-tracheostomy period44. Prophylactic antimicrobials have also been investigated in the context of recurrent LRTI34,36and persistent bacterial colonisation after LRTI33. Indeed, the most recent British Thoracic Society guidelines support the use of both enteral and inhaled antibiotics to reduce infection frequency in individuals with learning disabilities suffering from recurrent community acquired pneumonia45. Although early evidence on prophylactic inhaled antibiotic use in children with tracheostomies is promising, suggesting reduced frequency of LRTIs and systemic antimicrobial use, it is very weak, limited to two small retrospective cohort studies 30,31. Given the already troubling prevalence of multidrug resistant organisms in this population23, randomised controlled trial evidence is needed to ensure this common practice is actually beneficial and not detrimental to patients’ health.
Tracheal cultures are commonly obtained from paediatric tracheostomy patients, with the aim of guiding subsequent antimicrobial therapy. However, a positive culture alone is not diagnostic of infection. Positive cultures may represent normal respiratory organisms or colonisation of the respiratory tract, which is common after tracheostomy. Yalamanchi et al. highlight this challenge of knowing which culture results to act upon, reinforcing the need to interpret culture results as part of a wider clinical assessment39. Whilst Prinzi et al. demonstrated the potential harms of over-reporting aspirate cultures, which can lead to unnecessary antimicrobial exposure.37Further, Cline et al. demonstrate the danger of utilising historical culture results to guide current treatment, which they estimate would only be effective in half of cases.38The increasing uptake of next generation sequencing techniques, which are currently mainly limited to the research environment, may offer hope for the future. Comprehensive profiling of microbial communities will help us better understand the dynamic interplay between the tracheal microbiome and host-immune system.46,47In turn, improved understanding of what factors predispose individuals to microbial dysfunction and subsequent infection could provide alternatives to antimicrobial therapy to break the cycle of recurrent infections25.
The data summarised here have four main overriding limitations. Firstly, the populations studied were heterogeneous in terms of design, intervention, and outcome measures, limiting their generalisability. Secondly, there was no standardised definition of tracheostomy-associated respiratory infections across different studies, which often just reported clinical diagnosis, limiting their comparability. Features of interest when comparing datasets include the presence of positive tracheal aspirate cultures, radiographic changes, changes in gas exchange, and other clinical features (such as cough, sputum production, and pyrexia). Attention should be paid to infection definition when designing future studies to support interpretability of their results and subsequent meta-analysis. Thirdly, the small sample size of these studies precludes subgroup analysis, compounding the first two issues. Finally, all studies in our systematic review are retrospective cohort studies, which by design provide low level evidence for the efficacy of antimicrobial interventions. Prospective, randomised, controlled studies are needed to guide clinical decision making for this vulnerable patient cohort.