Description of the condition
Cystic fibrosis (CF) is the most common life-limiting autosomal recessive genetic disorder in Caucasians (Kosorok 1996). Respiratory disease is the major cause of mortality and morbidity in CF (Penketh 1987). About 40 years ago, most people with CF died in their first decade of life. With advances in treatment, life expectancy has increased dramatically since then from a median of two years to a projected median survival of over 50 years of age (Dodge 2007). One of the major reasons for this increase in survival is the mounting use of antibiotics to treat chest exacerbations caused by bacterial infections.
The abnormal CF gene causes a combination of increased sodium and defective chloride absorption across the airway epithelium, which in turn causes dehydration and leads to a build up of thick sticky mucus. This accumulates in the lungs and causes chronic pulmonary infections and bronchiectasis. In people with CF, bacterial pathogens are the main causes of respiratory exacerbations (Petersen 1981; Wat 2003).
In small children, Staphylococcus aureus (S. aureus) and Haemophilus influenzae (H. influenzae) are the most common organisms isolated and thought to be responsible for chest exacerbations, but pneumococci and sometimes Enterobacteriaceae are also isolated. In older children and adults, the major pathogen is Pseudomonas aeruginosa (P. aeruginosa), although the other organisms mentioned above can also be responsible and, in some centres Burkholderia cepacia (B. cepacia) too (Hodson 2000a).
At birth the lungs of infants with CF are normal, but shortly thereafter many become infected with bacteria (Accurso 1997). Endobronchial infection commonly occurs within the first one to two years of life (Armstrong 1995). The most common organism isolated from the sputum of adults is P. aeruginosa (Horre 2004), with traditionally 80% of people with CF being infected by 18 years of age (CF Foundation 2001; Hodson 2000b). This is no longer the case and chronic colonisation with P. aeruginosa can be prevented or delayed by segregation and aggressive eradication treatment of first isolates (Hansen 2008; Stuart 2010). Prevalence data from the UK indicates a rate of chronic P. aeruginosa infection of 45% in 16 to 19 year olds, riding to a peak of 67% in 28 to 31 year olds (CF Trust 2009). Chronic infection with P. aeruginosa is however associated with a more rapid decline in lung function (Emerson 2002; Schaedel 2002).
Most people with CF are initially infected with non-mucoid P. aeruginosa, which is relatively sensitive to antibiotics, but after a period of time it changes to a mucoid state (Starner 2005). Early acquisition of mucoid P. aeruginosa was associated with a four-fold greater decrease in cumulative survival (Demko 1995). Mucoid P. aeruginosa is much more difficult to treat and eradicate because it adopts a defensive mode of growth that leads to creation of a surrounding biofilm (Hentzer 2001; Høiby 2000). Biofilms are communities of bacteria enclosed in a self-produced polysaccharide matrix and which are adherent to a surface. The biofilm protects P. aeruginosa from normal host defences and antibiotics. In addition, bacterial adherence to mucus is increased in CF which also contributes to difficulties in clearing it from the airways (Donaldson 2003).
Description of the intervention
At the time of initial infection, the majority of P. aeruginosa isolates are susceptible to commonly used antibiotics but, as individuals are exposed to repeated courses of antimicrobial therapies, they often develop drug resistance (Gilligan 1999). Multi-resistant P. aeruginosa is associated with more severe lung disease, more rapid decline in FEV
Intravenous antibiotics are given either as home intravenous antibiotic treatment (HIVAT) or in hospital. For the treatment of chest exacerbations, the choice of intravenous antibiotics is generally based on recent culture results of airway secretions. In infections caused by P. aeruginosa, a combination of aminoglycoside and beta lactam is preferred as it enhances the activity against target organisms (synergy) and decreases the risk of antibiotic resistance developing (Cheng 1996).
How the intervention might work
Deciding on an optimal combination of antibiotic regimens depends on many factors, namely the type of organisms, antibiotic sensitivity, allergies, previous response to treatment and local policies. A good response to treatment can be expected if a suitable antibiotic combination is chosen.
The duration of a course of intravenous therapy to treat chest exacerbations varies but these are frequently between 10 and 21 days of antibiotics (CF Trust 2009; Doring 2000; Döring 2012; Gibson 2003; Hodson 2000a). The optimal duration of intravenous antibiotic therapy is not clearly defined (Rosenberg 1993). Response to therapy (as reflected by improved pulmonary function, oxygen saturation returning to pre-exacerbation levels, levelling off of weight loss, normalisation or significant falls in inflammatory markers, decreased bacterial density in sputum, improved well-being and clinical symptomatology scores) may be used to guide the optimal duration of intravenous antibiotic therapy (CF Trust 2009; Doring 2000; Döring 2012; Hodson 2000a; Ramsey 1996).
Individuals are usually commenced on a 10- to 14-day course of intravenous antibiotics (CF Trust 2009; Doring 2000; Döring 2012). This may be extended in people with severe exacerbations and incomplete recovery (Doring 2000; Döring 2012). However a consensus of opinion states that intravenous antibiotics should not be extended more than three weeks, except under very special circumstances and that patients with multi-drug resistant P. aeruginosa may require therapy (Döring 2012).Some studies also mention 10 days (CF Trust 2009; Doring 2000; Hodson 2000b; McLaughlin 1983) and 21 days (Gibson 2003; Ramsey 1996) as the duration required to treat chest exacerbations. With a shorter duration of antibiotic treatment there is a risk of inadequate clearance of infection which could lead to further lung damage. Prolonged courses of intravenous antibiotics are expensive and inconvenient to individuals. The incidence of allergic reactions to antibiotics also increases with prolonged courses (Koch 1991; Parmar 2005). Moreover, the use of aminoglycosides requires frequent monitoring of antibiotic levels to avoid some of their side effects namely, ototoxicity and nephrotoxicity (Hodson 2000b).
Why it is important to do this review
The aim of this review is to assess the available evidence in order to establish the optimal duration of intravenous antibiotic therapy for treating a chest exacerbation in people with CF.
To assess the optimal duration of intravenous antibiotic therapy for treating chest exacerbations in people with cystic fibrosis.
Criteria for considering studies for this review
Types of studies
Randomised controlled trials (RCTs) and quasi-randomised controlled trials.
Types of participants
People with CF, diagnosed clinically and by sweat or genetic testing, of all ages and all degrees of disease severity and who are being treated with intravenous antibiotics for a chest exacerbation.
We will require all studies to clearly state how they will define a chest exacerbation. We will regard as important or as a minimum, according to criteria published by the 1994 Cystic Fibrosis Foundation Microbiology and Infectious Disease Consensus Conference (CF Foundation 1994), the presence of at least three of the following eleven new findings or changes in clinical status when compared to the most recent baseline visit:
- increased cough;
- increased sputum production, change in appearance of expectorated sputum, or both;
- fever (greater or equal to 38°C for at least four hours in a 24-hour period) on more than one occasion in the previous week;
- weight loss greater than or equal to 1 kg or 5% of bodyweight associated with anorexia and decreased dietary intake;
- school or work absenteeism (due to illness) in the previous week;
- increased respiratory rate, increased work of breathing, or both;
- new finding on chest examination (e.g. rales, wheezing, crackles);
- decreased exercise tolerance;
- decrease in FEV
1of greater than or equal to 10% from previous baseline study within past three months;
- decrease in haemoglobin saturation (as measured by oximetry) from baseline value within past three months of greater than or equal to 10%;
- new finding on chest radiograph.
For paediatric chest exacerbations, the presence of three or more of the following characteristics will be used to define a chest exacerbation:
- in children under six years of age - new crackles, increased cough frequency, decline in weight and impression of increased sputum production;
- in children six years or over - relative decrease in per cent predicted FEV
1, increased cough frequency, new crackles and haemoptysis (Rabin 2004).
For the purpose of this study we will exclude people with CF who are treated electively with intravenous antibiotics and eradication regimens and only consider people with CF being treated for acute chest exacerbations.
Types of interventions
Comparison of different durations of intravenous antibiotic courses, which can either be the same drugs at the same dosage, the same drugs at a different dosage or frequency or different antibiotics altogether. We will not exclude studies in which there are additional therapeutic agents included e.g. prednisolone, azithromycin, dornase alpha or nebulised tobramycin or colistin.
Types of outcome measures
- Lung function (absolute change or percentage change compared to baseline values or both)
- forced expiratory volume at one second (FEV
1) (measured in L/min or % predicted)
- forced vital capacity (FVC) (measured in L/min or % predicted)
- expiratory flow from 25% to 75% of vital capacity (FEF
- Change in sputum bacteriology (quantitative e.g. colony-forming units per ml, or qualitative e.g. type of bacteria, or both)
- Adverse effects (e.g. allergic reactions, candidal infections)
- Quality of life (measured by health status questionnaires, if available)
- Change in nutritional status (absolute change or percentage change or both compared with baseline values) - Post hoc change: ideally, the following measures should be age-adjusted in paediatric participants
- body mass index (BMI)
- Time to next exacerbation
- Change in inflammatory markers
- measured in the sputum
- measured in the blood
- Treatment failure (defined as need to change to a different course of antibiotics because of clinical deterioration of the participant)
- Frequency of exacerbations of chest disease
- Lung function (absolute change or percentage change compared to baseline values or both)
1(measured in L/min or % predicted)
- FVC (measured in L/min or % predicted)
- Development of antibiotic-resistant strains
- Pseudomonas aeruginosa
- other organisms
- Adverse effects (e.g. decline in renal function)
- Number of intravenous antibiotic courses since study
- Quality of life (measured by health status questionnaires, if available)
Search methods for identification of studies
We identified relevant trials from the Group's Cystic Fibrosis Trials Register using the terms: antibiotics AND (intravenous OR not stated).
The Cystic Fibrosis Trials Register is compiled from electronic searches of the Cochrane Central Register of Controlled Trials (Clinical Trials) (updated each new issue of The Cochrane Library), quarterly searches of MEDLINE, a search of EMBASE to 1995 and the prospective handsearching of two journals - Pediatric Pulmonology and the Journal of Cystic Fibrosis. Unpublished work is identified by searching the abstract books of three major cystic fibrosis conferences: the International Cystic Fibrosis Conference; the European Cystic Fibrosis Conference and the North American Cystic Fibrosis Conference. For full details of all searching activities for the register, please see the relevant sections of the Cystic Fibrosis and Genetic Disorders Group Module.
Date of most recent search of the Cystic Fibrosis Trials Register: 15 November 2012.
Data collection and analysis
We were unable to identify any trials eligible for inclusion in this review. If we identify relevant trials for a future update we plan to undertake the following.
Selection of studies
Two authors (AP, MW) will independently apply the inclusion criteria to all potential trials. We will perform this without blinding. Any discrepancy that occurs in the trial selection will be resolved by discussion.
Data extraction and management
Two authors (AP, MW) will independently extract the data (using a customised data extraction form) and assess the methodological quality of the selected trials. Any discrepancy that occurs in the trial selection will be resolved by discussion.
For short-term outcomes, we plan to measure outcomes at less than a week, one to two weeks, more than two weeks to three weeks, more than three weeks to four weeks. We will also consider additional follow-up data recorded at other time periods.
Long-term outcomes will be measured from six months following the end of the course of antibiotics in the study. For long-term outcomes, we plan to measure outcomes at one month, up to three months, up to six months, up to twelve months and then annually thereafter. If outcome data are recorded at other time periods we will consider examining these as well.
Assessment of risk of bias in included studies
The risk of bias will be assessed using The Cochrane Collaboration "risk of bias" tool (Higgins 2011), The risk of bias will be considered for the following sections.
- Sequence generation: a detailed description of the method of used to generate the sequence allocation will be obtained and evaluated for the ability to produce bias.
- Allocation concealment: a detailed description of the methods used will be obtained. The ability to predict the randomisation of the current or future patients will be assessed.
- Blinding: as any eligible studies will be randomised controlled trials of duration of intravenous antibiotics the patients and clinicians will be aware of the duration of treatment unless placebo medicines are given for the additional days. Given the nature of the drugs and the effects it is likely that these could be detected by patients and healthcare professionals. The ability of any methods used to reduce bias will be considered in the following groups, patient, clinician and outcome assessor.
- Incomplete outcome data: we will review incomplete data for each main outcome for both attrition and exclusions. The studies will be assessed for an intention-to-treat analysis.
- Selective reporting: the outcomes will be reviewed for completeness of presentation. Any suggestions of selective reporting will be followed up with the authors.
- Other potential sources of bias: any other potential sources of bias will be considered according to the questions the study is answering.
Measures of treatment effect
For binary outcome measures, in order to allow an intention-to-treat analysis, we will seek data on the number of participants with each outcome event, by allocated treated group, irrespective of compliance and whether or not the individual was later thought to be ineligible or otherwise excluded from treatment or follow up. We aim to calculate a pooled estimate of the treatment effect for each outcome across studies using relative risk where appropriate.
For continuous outcomes, we plan to record either mean relative change from baseline for each group or mean post-treatment or intervention values and standard deviation. If standard errors are reported, and where possible, these will be converted to standard deviations. We will calculate a pooled estimate of treatment effect by calculating the weighted mean difference.
For any time-to-event outcomes included in the review, we plan to obtain a mixture of logrank and Cox model estimates from the trials; we will combine all results using the generic inverse variance method since we will be able to convert the logrank estimates into log hazard ratios and standard errors as detailed in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).
We will analyse longitudinal data using the most appropriate method available.
Unit of analysis issues
Ideally, when conducting a meta-analysis combining results from cross-over trials we will use the inverse variance methods that are recommended by Elbourne (Elbourne 2002). However, if there are restrictions on the data available, we may only be able to either use the first arm data only or to treat the cross-over trials as if they are parallel trials. Elbourne says that this approach will produce conservative results as it does not take into account within-patient correlation (Elbourne 2002). Also each participant will appear in both the treatment and control group, so the two groups will not be independent.
Dealing with missing data
We will contact the authors of the original studies for data for any missing individuals, outcomes or summary data. We plan to describe the implications of any missing data in detail and include this in the discussion. We may impute missing standard deviations according to the methodology described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We will collect any outcomes not available and report these in a risk of bias table. We plan to conduct an intention-to-treat analysis by contacting all authors for missing data. If there are large amounts of missing data which we can not impute with more than minimal assumptions, we will perform an available case analysis.
Assessment of heterogeneity
We plan to quantify the impact of statistical and clinical heterogeneity in the meta-analysis using a measure of the degree of inconsistency in the studies' results (Higgins 2003). This measure (I
- 0 to 40% might not be important;
- 30 to 60% may represent moderate heterogeneity;
- 50 to 90% may represent substantial heterogeneity;
- 75 to 100% represents considerable heterogeneity.
Assessment of reporting biases
We will attempt to assess whether our review is subject to publication bias by using a funnel plot. If asymmetry is detected, causes other than publication bias will be explored.
If no significant heterogeneity is identified, we will compute pooled estimates of the treatment effect for each outcome under a fixed-effect model.
Subgroup analysis and investigation of heterogeneity
If we find significant heterogeneity, we will investigate the possible causes of this further by exploring the impact of methodological quality and condition of the individuals (i.e. severity of disease, type of treatment, e.g. single or combined treatment) using subgroup analysis.
We will perform a subgroup analysis based on bacteriology - chronically infected with P. aeruginosa only, P. aeruginosa and other organisms (excluding B. cepacia), B. cepacia (with or without co-infection with other organisms) and non-P. aeruginosa organisms (excluding B. cepacia). Chronic infection is defined as at least two positive sputum cultures in the previous six months (CF Trust 2009).
We will compare duration of treatment used in HIVAT and hospital treatment separately. However, comparison of home versus hospital intravenous antibiotic treatment is the subject of another Cochrane review (Balaguer 2008) and this comparison is included here only as part of subgroup analysis to assess its impact on duration of treatment.
We will also compare duration of treatment based on age separately in three groups: under six years of age; between six and sixteen years of age; and over sixteen years of age.
We plan to perform a sensitivity analysis based on the methodological quality of the trials, including and excluding quasi-randomised trials.
Description of studies
No trials were identified which were eligible for inclusion in this review. A total of 225 papers were obtained from the Group's Cystic Fibrosis Trials Register. Out of these, only 82 papers involved treatment of chest exacerbations with intravenous antibiotics. The rest of the trials were excluded because they involved oral or nebulised antibiotics or were pharmacokinetic studies, studies concerning method of delivery or without any clinical outcome measures. Of the 82 papers that were reviewed, none of these RCTs compared different durations of antibiotic courses and are therefore not eligible for inclusion in this review.
Risk of bias in included studies
No trials were identified which were eligible for inclusion in this review.
Effects of interventions
No trials were identified which were eligible for inclusion in this review.
Cystic fibrosis is a life-limiting disease. While every effort needs to be focused on treating chest infections aggressively, there have been no clear guidelines on how long the optimum duration of intravenous antibiotics should be. Most studies show chest exacerbations are treated for 14 days as a routine (Aaron 2005; Burkhardt 2006; Hoogkamp 1983; Hyatt 1981; Smyth 2005; Vic 1998); however, there are studies which show 10 days are sufficient (Beaudry 1980; Master 2001; McLaughlin 1983; Penketh 1983; Penketh 1984). A study by Mendelman found 21 days to be more effective in treating chest exacerbations (Mendelman 1985). None of these studies are RCTs which compare different durations of antibiotic courses and are therefore not eligible to be included in the Results section of this review. There are no clear reasons for choosing 14 days or 21 days as the optimum duration, besides the assumption that longer treatment may clear the infection more thoroughly. Moreover, not all chest exacerbations are due to the usual bacteria known to infect the lungs of people with CF. Viruses, atypical bacteria and fungi are also implicated in causing exacerbations (Olesen 2006). Physiotherapy and bronchodilators on their own, sometimes, have the same effect as intravenous antibiotics in treating chest exacerbations (Gold 1987). As it is difficult to clinically distinguish bacterial and viral exacerbations, clinicians usually err on the side of treating them with antibiotics.
In most adults and adolescents with CF, eradication of the infecting bacterial organism is almost impossible as infections tend to be chronic and ineradicable. Therefore, extending duration of antibiotic therapy would not be expected to result in clearance of infection. In contrast to acute infections associated with other diseases, the duration of antibiotic therapy for CF exacerbations should probably be guided based on improvement in clinical status of the affected individual rather than based on attempts at rendering the airways sterile of bacteria.
We did not find any RCTs comparing different treatment durations. Reducing the treatment duration to 10 days may have significant benefits; improved quality of life, improved compliance, less incidence of drug reactions and lower cost. However, it is uncertain if this duration of 10 days is sufficient to clear a chest infection and does not result in an early recurrence of next exacerbation. Moreover, multi-resistant organisms may require longer duration to treat them effectively.
Implications for practice
There are no published data to recommend the optimum duration of intravenous antibiotic therapy for treating a chest exacerbation. Duration of treatment is currently decided based on unit policies and the individual's response to treatment and we have found no evidence to change this practice.
Implications for research
This systematic review has identified the need for a well-designed, adequately-powered, multicentre randomised controlled trial to assess the optimum duration of intravenous antibiotic therapy to treat chest exacerbations, an issue which has important clinical and financial implications. This type of study could be done by randomising participants to receive either 7 days, 10 days, 14 days or 21 days of intravenous antibiotics and comparing outcome measures at these various end points. Long-term follow up is needed to look at the time to next exacerbation, frequency of chest exacerbations and development of antibiotic-resistant strains. Outcome measures, such as pulmonary function tests, time to next exacerbation and improvement in quality of life scores should be clearly stated and multicentre trials should be co-ordinated so that maximum power is available to achieve a clear result.
To Dr Frank Edenborough for his advice and clinical perspective and to Mrs Nikki Jahnke from The Cochrane Cystic Fibrosis & Genetic Disorders Group for her help and editorial advice during preparation of this review.
2010 - the current authors (AP and MW) would like to thank Dr Bryan Fernandes for his input into previous versions of this review.
Data and analyses
This review has no analyses.
Last assessed as up-to-date: 27 March 2013.
Protocol first published: Issue 3, 2007
Review first published: Issue 2, 2008
Contributions of authors
Declarations of interest
At the update in 2010 Dr Bryan Fernandes left the review team and Amanda Plummer took on the role of lead author.
Medical Subject Headings (MeSH)
Anti-Bacterial Agents [*administration & dosage]; Bacterial Infections [*drug therapy]; Cystic Fibrosis [*complications]; Drug Administration Schedule; Injections, Intravenous; Lung Diseases [*drug therapy]
MeSH check words