Pulmonary exacerbation in adults with bronchiectasis: a consensus definition for clinical research

Discussion

This consensus definition of exacerbation may serve to standardise a key outcome measure of clinical trials involving adults with bronchiectasis and allow more meaningful comparisons of treatment effect between interventions. Although the remit of the working group was to define pulmonary exacerbation in adults with bronchiectasis in the context of prospective clinical trials, it is anticipated that this definition will inform patients and clinicians to more accurately identify exacerbations. We acknowledge that this definition is not a validated instrument, which might be more useful in determining if treatment of an exacerbation is efficacious. However, it does provide an agreed-upon definition by a group of bronchiectasis experts using a systematic approach intended to define a clinical end-point in the treatment of bronchiectasis and represents a proposed definition to be used for validation in future investigations.

The systematic review performed prior to the working group meeting provided an evidence-based platform for the factors included in the Delphi process. The subsequent outputs then enabled investigators to discuss and establish arguments so as to agree on the key components of the definition.

The consensus definition has three core components: collection of symptoms, time and decision to change treatment. Owing to the lack of evidence and validation data, any recommended criteria are preliminary and may be modified as more data become available.

Previous definitions used in trials had included a wide variety of symptoms but only six made the final list for this proposed definition. For some criteria it was deemed they were not related to exacerbation of lower airways disease (e.g. sinus disease and heart failure), while for others the criteria were felt to be infrequently measured in clinical practice to be relevant (e.g. CRP, serum IL-6 and serum amyloid) [1, 25, 26]. Spirometric measures were discounted because, unlike in CF, changes in FEV₁ and FVC are not reliable indicators of exacerbation or response to treatment in patients with bronchiectasis [25]. Auscultatory findings were also felt to not be sufficiently bronchiectasis specific and are largely dependent on the severity of the predominant underlying lung condition. Chest radiograph changes were excluded as it can be difficult to distinguish pneumonia from mucus plugging, a finding that is common in bronchiectasis, and because most exacerbations occur in outpatient settings without access to radiology. A further important consideration during definition development was the lack of availability of some factors in domiciliary and certain healthcare settings. A symptom-based definition was therefore deemed preferable.

There was considerable debate regarding the requirement of a minimum number of criteria within the definition. It is assumed that a greater number of criteria met would increase the specificity of the diagnosis, and a consensus view that three out of six criteria be met was based upon a desire to achieve sufficient balance between sensitivity and specificity for use in clinical trials. This does not suggest that a patient cannot have an exacerbation when meeting only two out of six criteria. Nor is it known that such exacerbations are the types we wish to prevent and future trials validating the number of symptoms that best define a diagnosis of exacerbation are needed.

Time was included in the definition to address the recognised day-to-day variability of symptoms in people with bronchiectasis. The international group recognised some patients feel worse over a 24-h period and then spontaneously improve without the need for augmentation or change of their treatment. Symptom duration of ≥48 h to be certain there was a real change in the condition was considered more appropriate than a threshold of only 24 h. This element of the definition was thought to reduce inappropriate antibiotic prescribing and support adherence to the principal of antibiotic stewardship. If, however, there is a severe attack within 48 h necessitating treatment, such patients would still qualify with the proposed definition as they will have symptoms lasting ≥48 h, as the treatment takes time to improve symptoms.

Finally, there is the inclusion of a need for intervention. Although most exacerbations are treated with antibiotics, the committee felt that not all exacerbations require antibiotic therapy as they could be self-limiting (e.g. viral illness) or respond to alternate therapy (e.g. steroids for allergic bronchopulmonary aspergillosis). Studies of specific medications may elect to define the specific intervention. Symptoms of sufficient severity warranting a change in bronchiectasis treatment were also thought to represent a minimum threshold for definition of exacerbation for potential intervention and purposes of development of a definition for clinical intervention trials. Formal grading of the severity of exacerbation into mild, moderate and severe was not felt to be critical for the objective of this work.

It is important to capture all exacerbations in which a physician makes a decision to treat as part of the outcomes in clinical trials. The purpose of a definition is not to override the clinical judgement of physicians but to achieve some objective verification of that decision in multicentre, often international trials, where the threshold for that decision to treat is variable.

In conclusion, the working group comprising bronchiectasis experts from Europe (representing EMBARC), North America (representing BRR), Australasia and South Africa proposes the use of this consensus-based definition for bronchiectasis exacerbation (figure 2) in clinical trials involving adults with bronchiectasis. Although there was no representation from Asia, we believe this definition will be applicable internationally but will need to be validated in different settings. Although the evidence base was limited, international experts propose this workable definition and the plan is that there will be validation of this definition in future trials.