Historical cohort, longitudinal, observational, and multicenter study (43 centers from all over Spain) of a cohort of adult patients (at least 18 years) derived from the Spanish Bronchiectasis Registry (RIBRON) [14] and diagnosed with bronchiectasis by high-resolution computed tomography, presenting compatible clinical symptoms, and followed up at a hospital center.

Baseline data were available on 2630 patients included from February 2015 to December 2020 and mortality data were available from February 2015 to June 2023. A wide range of baseline inclusion criteria was available in a clinical stable situation: complete general, clinical, analytical, functional, radiological, microbiological, and therapeutic data available at entry, conditions of clinical stability (defined as at least 4 weeks free of an exacerbation period), and completed data to construct the multidimensional severity scores (FACED, E-FACED, BSI [11–13]). FACED is the acromyn of FEV1, Age, Chronic colonization by P. aeruginosa Extension (radiological, in number of pulmonary lobes affected) and Dyspnea (measure by Modified Medical Research Council); E-FACED is identical to FACED, with the addition of the number of exacerbations in the previous year. Finally, BSI consists of eight variables: Age, body mass index, FEV1% predicted, hospital admissions and exacerbations before study, dyspnea, PA colonization, colonization by other organisms and radiological severity

There was a strict control of the quality of data as soon as they were entered, with alarms set to identify values that were included abnormally, or not included at all (despite being requested from the centers). The main criteria for exclusion were a diagnosis of cystic fibrosis (CF) and a lack of information on mortality.

The local ethics committee affiliated with the registry (number: 001-2012. Hospital Josep Trueta, Girona) approved the study and all the patients gave their written informed consent in their corresponding participating center.

Starting with the data collected closest to the time of the patient's inclusion in the registry, any data on changes in a patient's clinical-functional characteristics, new microbiological and analytical factors, and any important occurrences were recorded prospectively on a yearly basis. Data on exacerbations during the year prior to the time of the patient's inclusion in the study were also recorded.

Exacerbation was defined as a worsening of the typical symptoms of bronchiectasis: cough; increase in the purulence or volume of the sputum; dyspnea; hemoptysis; and chest pain with an evolution of more than 24h for which antibiotic treatment was required. In terms of severity, we classified the exacerbations into two subgroups: mild-moderate, if the patients needed oral antibiotics; and severe (hospitalizations), in cases where hospital admittance or intravenous antibiotic treatment was required, even if the latter was administered at home [15]. CBI was defined as the presence of three or more consecutive cultures positive for the same potentially pathological microorganism [16]. Neutrophilia was defined as the presence of at least 80% o neutrophils in a peripheral blood sample [17]. Diagnosis delay was defined as the time elapse between the onset of symptoms and the diagnosis of bronquiectasis

Vital status at the end of follow-up was thoroughly assessed using multiple concurrent approaches, including computerized databases reviewing hospital's medical outpatient records and, when necessary, by contacting the patient, patient's relatives, or primary care physician. When a participant died, information about the cause and date of death was obtained from official death certificates or from the hospital medical record, if the patient died in hospital.

Continuous variables were presented as mean (standard deviation [SD]) or median (interquartile range), depending on the distribution of the variables. Categorical variables were presented as relative frequency and percentage. The Kolmogorov–Smirnov test was used to assess the normal distribution of the variables. Comparison between two groups (patients who died versus patients who did not) was assessed using the t-Student, Mann–Whitney, or Chi-square test, depending on the distribution and type of the variables. The Pearson coefficient was used to calculate correlations between variables.

Cumulative all-cause mortality for each studied group was calculated according to the Kaplan–Meier method, and mortality curves were compared by log-rank test. Censoring was applied to individuals who did not died by the end of the follow-up period. Variables that were clinically relevant, in the opinion of the researchers, were entered into a Cox proportional hazard model analysis to determine the variables independently associated with mortality in the study groups. The following variables were finally selected to enter in the Cox model: age, gender, diagnosis delay, sputum purulence (Murray scale), dyspnea, comorbidities (Charlson Index), presence of CBI, pathogen microorganisms isolate, previous cardiovascular events, arterial hypertension, diabetes mellitus, FEV1%postbronchodilator, etiology, presence of COPD or asthma, CT extension of bronchiectasis (number of pulmonary lobes affected), treatments (inhaled corticosteroids, long-acting bronchodilator use, use of macrolides and inhaled antibiotics), as well as analytical data (absolute number and percentage): neutrophils, lymphocytes, eosinophils, platelets, fibrinogen (mg/ml), and C-reactive protein (IU/mL). Multidimensional severity scores were analyzed separately since they were formed by the combination of some of the previous variables. Diagnostic and residual plots were examined for all variables to test the proportional hazards assumptions, and none of them were statistically significant.

The results were expressed as hazard ratio (HR) and 95% confidence interval (CI) as well as AUC-ROC, with a p value less than 0.05 considered as statistically significant. The SPSS 22.0 statistical package (SPSS Inc., Chicago, IL) was used for the analysis.

Table 2 shows the comparison of baseline and follow-up variables between those patients alive and dead at the end of the study. There were 408 patient deaths during follow-up (19.8%). The deceased patients generally presented more severe bronchiectasis in almost all the domains that describe the disorder, as well as a greater number of comorbidities. Patients were followed up for a median (interquartile range) of 84 (72–102) months. The causes of death were respiratory (n=201; 49.3%), cardiovascular (n=48; 11.2%), malignant (n=44; 10.8%), and others or unknown (n=115, 28.7%).

Thirteen variables were independently associated with mortality, as can be seen in Fig. 2, Table 1 Suppl. The strongest association with bronchiectasis mortality was the concomitant presence of COPD (HR 3.16; 95%CI: 2.5–3.9), the presence of CBI by PA (HR 2.57; 95%CI: 2.88–2.21), previous and follow-up hospitalizations (HR 2.01; 95%CI: 2.59–1.88 and HR 1.93; 95%CI: 1.63–2.32, respectively), as well as the presence of neutrophilia (HR 2.53; 95%CI: 1.81–3.61). Other variables such as baseline and follow-up exacerbations, other comorbidities, age, FEV1, dyspnea, and pneumonic episodes were also associated with higher bronchiectasis mortality. Only the presence of baseline peripheral neutrophilia appears as a new factor associated to mortality in bronchiectasis. When a separate analysis was performed for deaths of respiratory and cardiac origin, we observed that the three factors most strongly associated with respiratory mortality were the presence of COPD (HR 3.27; 95% CI: 2.53–3.91), chronic bronchial infection by P. aeruginosa (HR 2.42; 95% CI: 2.04–3.27), and neutrophilia>80% (HR 2.30; 95% CI: 1.21–2.41). In contrast, for cardiovascular deaths, the main associated factors were previous cardiovascular events (HR 4.17; 95% CI: 4.01–5.74), arterial hypertension (HR 3.15; 95% CI: 2.71–4.09), and diabetes mellitus (HR 2.17; 95% CI: 2.61–4.11). Further details of these analyses are presented in Supplementary Table 2.

Fig. 2.

Concentric graphic. Independent risk factor for mortality in bronchiectasis.

Fig. 3A shows a more detailed analysis of the exacerbations. There is a risk of greater mortality from mild or moderate exacerbations when there are two previous exacerbations or at least two exacerbations per year during the follow up, while, as regards hospitalizations (Fig. 3B), the presence of just one hospitalization previously or during the follow-up is sufficient to increase this risk by 2.60 and 1.83, respectively.

Fig. 3.

Baseline and follow-up of mild-to-moderate (A) and severe exacerbations (hospitalization) (B) and their relationship with bronchiectasis mortality.

Both the presence of chronic mucopurulent or purulent sputum and a volume>20 cc per day were associated with higher mortality in bronchiectasis (Fig. 1 suppl).

On the other hand, the presence of CBI by PA was a strong factor in the relationship with mortality (HR 2.57; 96%CI: 2.88–2.21) as it has been in other studies (Fig. 2 suppl). However the coexistence of COPD was the strongest variable associated with long-term mortality in bronchiectasis, with an HR greater than three (3.16; 95%CI: 2.54–3.88) also seen in other previous studies (Fig. 3 suppl.)