Abstract
Introduction
Pseudomonas aeruginosa (PsA) infection contributes to disease progression in bronchiectasis (BE), particularly exacerbations which are known to increase the risk of cardiovascular (CV) events. However, the link between PsA infection and CV events in BE is unknown. Thus, we investigated whether there is an association between PsA airway infection and the risk of CV events post-exacerbation.
Methods
This was a US retrospective cohort study using the TriNetX platform between 2008 and 2019. Adult patients with (wPsA) or without (w/oPsA) PsA airway infection were included. Date of first exacerbation corresponded to the index date, and patients were followed for up to 5 years post index. Risk ratios (RR) for hospitalisation, subsequent exacerbation, mortality and incidence of pre-specified CV events were estimated. Propensity score matching (PSM) was used to balance baseline characteristics.
Results
After PSM, patients wPsA infection were at a greater risk of hospitalisation (RR: 1.40; 95% CI: 1.19–1.64), subsequent exacerbation (RR: 1.70; 95% CI: 1.53–1.90) and mortality (RR: 1.37; 95% CI: 1.20–1.56) than patients w/oPsA. PsA infection was associated with a higher risk of dysrhythmias (RR: 1.32; 95% CI: 1.13–1.54), inflammatory heart disease (RR: 2.09; 95% CI: 1.29–3.37), other cardiac disorders (RR: 1.40; 95% CI: 1.14–1.72), thrombotic disorders (RR: 1.31; 95% CI: 1.01–1.68), major adverse cardiovascular events (RR: 1.35; 95% CI: 1.19–1.52) and any CV outcome (RR: 1.42; 95% CI: 1.24–1.62).
Conclusion
PsA infection in patients with BE is associated with an increased risk of CV events following a baseline exacerbation. These data highlight the multisystemic nature of BE and the need to raise awareness of the potential increased risk of CV events in patients with BE experiencing exacerbations.
Shareable abstract
Patients with bronchiectasis with Pseudomonas aeruginosa airway infection have a higher risk of cardiovascular (CV) events post-exacerbation. Monitoring CV risk post-exacerbation in these patients should be considered. https://bit.ly/42wgIhp
Introduction
Bronchiectasis (BE) is a chronic inflammatory pulmonary disorder characterised by permanent dilatation of the bronchi and is associated with cough, sputum production and exacerbations [1]. BE is associated with multiple comorbidities which contribute to the disease burden and mortality [2]. Cardiovascular diseases (CVD) are commonly reported in patients with BE [3, 4]. Previous studies have identified a higher prevalence of coronary artery disease and stroke in patients with BE versus the general population [5, 6]. A study in Spain reported that exacerbations further increase the risk of cardiovascular (CV) events [7]. Previous research has noted a time-dependent association between hospitalisation due to an infection and a subsequent abrupt increase in CV risk which appears to be inconsistent with temporal patterns for other risk factors related to CV events [8].
Respiratory pathogens may contribute to the pathogenesis of CV events following infection [9]. However, whether specific pathogens contribute to the increased CVD risk in patients with BE has not been studied and large longitudinal studies of CVD risk in patients with BE across different regions are lacking.
Pseudomonas aeruginosa (PsA), a gram-negative bacterium, is commonly responsible for chronic airway infection in patients with BE with an estimated prevalence of 25% and 15% in the US [10] and Europe [11], respectively. PsA infection in patients with BE is associated with poorer outcomes including greater lung function impairment, more frequent exacerbations, worse quality of life, greater risk of hospitalisations and increased mortality rates [12]. Whether patients with BE who are infected with PsA are at greater risk of CV events following an exacerbation than those who are not infected with PsA has not been explored.
Systemic inflammation, which is linked to the pathogenesis of atherosclerosis [13], is increased following an exacerbation of BE, and is increased to a greater extent and for longer in patients with chronic PsA airway infection versus those without PsA [14]. Therefore, this study tested the hypothesis that airway PsA infection is linked to an increased risk of CV events among BE patients with a history of exacerbation.
Methods
Data source
We leveraged the TriNetX LIVE database (TriNetX Inc., USA) to conduct this US-based retrospective study. TriNetX LIVE is a clinical research network which collects real-time electronic medical data from healthcare organisations. Data included demographics, diagnoses (based on International Classification of Diseases, 10th Revision (ICD-10) disease codes), medical services and procedures, medication details and laboratory measurements. As the network provides only aggregated counts and statistical summaries of deidentified information, this retrospective study is exempt from the requirement for informed consent.
Study design and population
Administrative claims data or electronic medical records (EMRs) are commonly used to conduct epidemiological studies in BE [15, 16]. We applied a published algorithm to identify our study population [17]. Adults (≥18 years) with BE were included in this analysis and entered the cohort from January 2008 to December 2014 (supplementary figure S1). The cohort entry date was defined as the date of first diagnosis of BE with a final date by December 2014. All patients were required to have ≥12 months of medical data available prior to the index date (ID). Baseline characteristics including demographics, comorbidities and the use of medications for managing BE and heart diseases were collected during the 12 months before ID. Patients were excluded from entering the cohort if they had a prior history of CVD. However, CV diseases and related medications after entry into the cohort and before the index event were included in the propensity score matching (PSM). The date of a patient's first BE exacerbation was used as the ID (see supplementary material for exacerbation definition). The follow-up period was up to 5 years post ID. Patients meeting the inclusion criteria at baseline were divided into two cohorts: those with PsA infection (wPsA) and those without PsA infection ever (w/oPsA) (supplementary figure S1). Supplementary table S1 shows the diagnosis and procedure codes used for patient selection. Two balanced cohorts were created by performing 1:1 PSM to reduce confounding (see supplementary material for more details). Supplementary table S2 lists covariates used in the PSM.
Study outcomes
Outcomes were captured during the follow-up period and identified based on ICD10 codes and included BE exacerbation, all-cause hospitalisation, CV events as described previously [18] and all-cause mortality. CV events were assessed on an individual and group basis. Major cardiovascular event (MACE) and a composite outcome of any pre-specified CV event were included in the CV assessment (supplementary table S3). In addition, we conducted sensitivity analyses to validate the reliability of our primary results. First, we included relevant laboratory data (lipid biomarkers and glycated haemoglobin) that had substantial missingness in the PSM process (supplementary table S2). Second, we calculated E-values to quantify the effect of unmeasured confounding (see supplementary material for more details).
Statistical analysis
All statistical analyses were performed in real-time using TriNetX. For each outcome, the risk difference was calculated with a 95% confidence interval (95% CI) and the z-test was used to evaluate statistical significance. Risk ratios (RR) with 95% CI were also calculated for all outcomes. Kaplan–Meier method was used for the survival probability analysis, and the log-rank test was used to evaluate statistical significance. Patients were censored at the end of the follow-up period or on the day after the last event in their health records. Significance was set as a p-value <0.05 (two-sided). The number of exacerbations in each cohort was assessed using the “number of instances” analysis tool provided by TriNetX. A t-test was performed on the number of instances of exacerbation per cohort.
Results
Characteristics of the cohort
A total of 998 patients with BE had a record of PsA infection and 9692 did not (figure 1). All baseline characteristics before and after PSM are described in supplementary table S4 and table 1, respectively. Before PSM, there were more women than men in both groups and the mean age at the ID was 63 and 64 years in the wPsA and w/oPsA groups, respectively. The prevalence of comorbidities that are associated with CV events including COPD, type 2 diabetes mellitus, hypertension and obesity was higher in the wPsA than the w/oPsA group. Patients wPsA had higher use of inhaled corticosteroids and bronchodilators, oral corticosteroids and antibiotics than those in the w/oPsA group. There was higher use of cardiovascular medications, except for angiotensin receptor blockers and antilipemic agents, in the wPsA versus the w/oPsA group (supplementary table S4). After PSM, 993 patients wPsA were matched successfully and there were minimal differences in baseline characteristics between the two groups (table 1). The mean follow-up was 1304±686 days and 1369±660 days for the wPsA and w/oPSA groups, respectively.
FIGURE 1.
Cohort selection flowchart. BE: bronchiectasis.
TABLE 1.
Baseline characteristics of patients after propensity score matching (1 year before index date)
| Characteristic | BE wPSA | BE w/oPsA | SMD |
|---|---|---|---|
| Subjects, n | 993 | 993 | |
| Age at index years, mean±sd | 63.5±17.1 | 63.8±17.5 | 0.021 |
| Sex, n (%) | |||
| Female | 584 (58.8) | 577 (58.1) | 0.014 |
| Male | 399 (40.2) | 409 (41.2) | 0.021 |
| Not reported | 10 (1) | 7 (0.7) | NA |
| Race, n (%) | |||
| White | 716 (72.1) | 711 (71.6) | 0.011 |
| African-American | 91 (9.2) | 94 (9.5) | 0.010 |
| Asian | 46 (4.6) | 46 (4.6) | <0.001 |
| Unknown | 118 (11.9) | 120 (12.1) | 0.006 |
| BMI, kg·m−2, mean±sd | 24.4±6 | 25.5±6.6 | 0.174 |
| Missing, n (%) | 453 (45.6) | 483 (48.6) | |
| Lifestyle, n (%) | |||
| Nicotine dependence | 52 (5.2) | 47 (4.7) | 0.023 |
| Tobacco use | 10 (1) | 10 (1) | <0.001 |
| Comorbidities, n (%) | |||
| Asthma | 165 (16.6) | 149 (15) | 0.044 |
| COPD | 276 (27.8) | 272 (27.4) | 0.009 |
| HTN | 346 (34.8) | 328 (33) | 0.038 |
| Dyslipidaemia | 216 (21.8) | 200 (20.1) | 0.040 |
| T2D | 142 (14.3) | 127 (12.8) | 0.044 |
| IHD | 119 (12) | 103 (10.4) | 0.051 |
| Obesity | 53 (5.3) | 44 (4.4) | 0.042 |
| CKD | 75 (7.6) | 81 (8.2) | 0.022 |
| Cough | 344 (34.6) | 325 (32.7) | 0.040 |
| Wheezing | 37 (3.7) | 39 (3.9) | 0.010 |
| Chest pain | 118 (11.9) | 112 (11.3) | 0.019 |
| Fatigue | 116 (11.7) | 104 (10.5) | 0.039 |
| Dyspnoea | 361 (36.4) | 336 (33.8) | 0.053 |
| Haemoptysis | 43 (4.3) | 28 (2.8) | 0.081 |
| sBP, mmHg, mean±sd | 121.9±21.7 | 123.9±20.5 | 0.096 |
| Missing, n (%) | 432 (43.5) | 461 (46.4) | |
| BE medication, n (%) | |||
| ICS | 382 (38.5) | 371 (37.4) | 0.023 |
| OCS | 470 (47.3) | 476 (47.9) | 0.012 |
| Bronchodilators | 569 (57.3) | 565 (56.9) | 0.008 |
| Quinolones | 361 (36.4) | 350 (35.2) | 0.023 |
| Macrolides | 254 (25.6) | 219 (22.1) | 0.083 |
| Blood and cardiac medication, n (%) | |||
| Anticoagulants | 315 (31.7) | 291 (29.3) | 0.053 |
| Antiarrhythmics | 227 (22.9) | 215 (21.7) | 0.029 |
| Antilipemic agents | 185 (18.6) | 161 (16.2) | 0.064 |
| β-blockers | 213 (21.5) | 201 (20.2) | 0.030 |
| CCBs | 178 (17.9) | 180 (18.1) | 0.005 |
| ACE inhibitors | 129 (13) | 117 (11.8) | 0.037 |
| ARBs | 58 (5.8) | 59 (5.9) | 0.004 |
| Digitalis | 19 (1.9) | 16 (1.6) | 0.023 |
| Antiplatelets | 186 (18.7) | 167 (16.8) | 0.050 |
| Diuretics | 245 (24.7) | 246 (24.8) | 0.002 |
Standardised mean differences: bold font represents a SMD >0.1 (i.e. variables not adequately matched). BE: bronchiectasis; wPsA: with Pseudomonas aeruginosa; w/oPsA: without P. aeruginosa; SMD: standardised mean difference; BMI: body mass index; HTN: hypertension; T2D: type 2 diabetes mellitus; IHD: ischaemic heart disease; CKD: chronic kidney disease; sBP: systolic blood pressure; ICS: inhaled corticosteroids; OCS: oral corticosteroids; CCBs: calcium channel blockers; ACE: angiotensin-converting enzyme; ARBs: angiotensin receptor blockers.
Exacerbation outcome
After PSM, the risk of subsequent exacerbations following the first occurrence of a BE exacerbation (the index BE exacerbation) was significantly higher in the wPsA group than the w/oPsA group (RR: 1.70; 95% CI: 1.53–1.90) in the 5-year follow-up period, with a median of three subsequent exacerbations in the wPsA group compared with two in the w/oPsA group (table 2, figure 2a; supplementary table S5 shows the RR before PSM). The cumulative probability at the end of the time window of subsequent exacerbations post index was significantly lower in the w/oPsA group compared with the wPsA group (37% versus 63%, log-rank p=0.0001) (figure 2b).
TABLE 2.
Risk of exacerbation, hospitalisation and all-cause mortality in patients with BE wPsA (n=993) compared with patients with BE w/oPsA (n=993)
| Outcomes | BE wPSA | BE w/oPsA | ARR (95% CI) | ARD % (95% CI) | p-value |
|---|---|---|---|---|---|
| Exacerbations | 542/993 | 318/993 | 1.70 (1.53–1.90) | 22.6 (18.3–26.8) | <0.0001 |
| Hospitalisation | 167/296 | 137/339 | 1.40 (1.19–1.64) | 16 (8.3–23.7) | <0.0001 |
| All-cause mortality | 360/981 | 264/985 | 1.37 (1.20–1.56) | 9.9 (5.8–14) | <0.0001 |
Propensity score matching applied. BE: bronchiectasis; wPsA: with Pseudomonas aeruginosa; w/oPsA: without P. aeruginosa; ARR: adjusted risk ratio; ARD: adjusted risk difference.
FIGURE 2.
a) Forest plot showing the risk ratio for hospitalisation, exacerbation and all-cause mortality in patients with bronchiectasis (BE) with Pseudomonas aeruginosa (wPsA) infection (n=993) compared with propensity-score matched patients with BE without PsA (w/oPsA) (n=993). Outcomes were followed in both cohorts for 5 years. Adjusted risk ratios and 95% CIs are presented. b) Kaplan–Meier curves of subsequent exacerbations post index exacerbation. c) Kaplan–Meier curves of hospitalisation post index exacerbation. d) Kaplan–Meier curves of all-cause mortality post index exacerbation.
Hospitalisation and all-cause mortality outcomes
Patients wPsA had a higher risk of hospitalisation post exacerbation than matched patients w/oPsA (RR: 1.40; 95% CI: 1.19–1.64; table 2 and figure 2a; supplementary table S5 shows the RR before PSM). The actual median time to a subsequent hospitalisation post exacerbation was significantly lower in the wPsA group compared with the matched w/oPsA group as shown by the Kaplan–Meier curve (85 versus 338 days, log-rank p=0.0001) (figure 2c).
Patients with BE in the wPsA group were 1.4-fold more likely to die than patients w/oPsA over the follow-up period (RR: 1.37; 95% CI: 1.20–1.56; table 2 and figure 2a; supplementary table S5 shows the RR before PSM). The survival probability over the 5-year follow-up period was also lower in the wPsA versus the matched w/oPsA group (59% versus 70%, log-rank p<0.0001, figure 2d).
Cardiovascular events
During the 5-year follow-up period post index exacerbation, the wPsA group had a higher RR for some CV events than the matched w/oPsA group (table 3 and figure 3a; supplementary table S6 shows the RR before the PSM). PsA was associated with higher RR for atrial fibrillation and flutter (RR: 1.37; 95% CI: 1.06–1.77), tachycardia (RR: 1.34; 95% CI: 1.09–1.64), bradycardia (RR: 1.47; 95% CI: 1.18–1.83), ventricular arrhythmias (RR: 1.31; 95% CI: 1.08–1.59), pericarditis (RR: 2.13; 95% CI: 1.31–3.47), heart failure (RR: 1.34; 95% CI: 1.08–1.67) and cardiac arrest (RR: 1.93; 95% CI: 1.04–3.58). There was also a non-significant trend towards higher RR for deep vein thrombosis (RR: 1.35; 95% CI: 0.96–1.89) and superficial vein thrombosis (RR: 1.47; 95% CI: 0.99–2.18) in the wPsA cohort. The RRs for composite CV events indicated that the risks were higher in the wPsA cohort than the w/oPsA group (table 4 and figure 3b; supplementary table S7 shows the RR before PSM). Specifically, the wPsA group was associated with a relative increase in dysrhythmias (RR: 1.32; 95% CI: 1.13–1.54), inflammatory heart disease (RR: 2.09; 95% CI: 1.29–3.37), other CV disorders (RR: 1.40; 95% CI: 1.14–1.72) and thrombotic disorders (RR: 1.31; 95% CI: 1.01–1.68). Analysis of MACE and any CV event revealed that patients wPsA were at a greater risk compared with patients w/oPsA in the 5-year follow-up period post index exacerbation. Notably, there was a 35% increase in the relative risk of MACE and a 42% increase in the relative risk of any CV event in the wPsA compared with the w/oPsA group (RR: 1.35; 95% CI: 1.19–1.52; RR: 1.42; 95% CI: 1.24–1.62; respectively). However, the statistical significance of the association between PsA and MACE was not maintained at the component level for myocardial infarction (RR: 1.16; 95% CI: 0.82–1.64) or stroke (RR: 1.14; 95% CI: 0.87–1.50). Similarly, PsA was not associated with a relative increase in transient ischaemic attack (RR: 1.14; 95% CI: 0.65–2.01) (table 3).
TABLE 3.
Risk of cardiovascular events in patients with BE wPsA (n=993) compared with patients with BE w/oPsA (n=993)
| Outcomes | BE wPSA | BE w/oPsA | ARR (95% CI) | ARD % (95% CI) | p-value |
|---|---|---|---|---|---|
| CV accidents | |||||
| Stroke | 100/899 | 88/902 | 1.14 (0.871.50) | 1.4 (−1.5–4.2) | 0.343 |
| TIA | 25/978 | 22/980 | 1.14 (0.65–2.01) | 0.3 (−1.0–1.7) | 0.653 |
| Dysrhythmias | |||||
| AF/AFL | 119/866 | 89/887 | 1.37 (1.06–1.77) | 3.7 (0.7–6.7) | 0.016 |
| Tachycardia | 179/868 | 137/889 | 1.34 (1.09–1.64) | 5.2 (1.6–8.8) | 0.005 |
| Bradycardia | 163/866 | 114/888 | 1.47 (1.18–1.83) | 6 (2.6–9.4) | <0.001 |
| VA | 193/835 | 149/845 | 1.31 (1.08–1.59) | 5.5 (1.6–9.3) | 0.005 |
| Inflammatory HD | |||||
| Pericarditis | 49/970 | 23/971 | 2.13 (1.31–3.47) | 2.7 (1–4.4) | 0.002 |
| Myocarditis | ≤10#/990 | ≤10#/991 | NA | NA | NA |
| IHD | |||||
| ACD | 17/983 | 11/987 | 1.55 (0.73–3.30) | 0.6 (−0.4–1.7) | 0.249 |
| MI | 65/958 | 57/976 | 1.16 (0.82–1.64) | 1 (−1.2–3.1) | 0.393 |
| ICM | ≤10#/980 | 11/984 | NA | NA | NA |
| Angina | 36/971 | 32/969 | 1.12 (0.70–1.79) | 0.4 (−1.2–2.0) | 0.628 |
| Other cardiac disorders | |||||
| HF | 161/849 | 119/842 | 1.34 (1.08–1.67) | 4.8 (1.3–8.4) | 0.008 |
| NICM | 45/948 | 45/954 | 1.01 (0.67–1.51) | 0 (−1.9–1.9) | 0.976 |
| Cardiac arrest | 29/986 | 15/986 | 1.93 (1.04–3.58) | 1.4 (0.1–2.7) | 0.033 |
| CS | 14/990 | ≤10#/989 | NA | NA | NA |
| Thrombotic disorders | |||||
| PE | 54/934 | 42/952 | 1.31 (0.89–1.94) | 1.4 (−0.6–3.4) | 0.176 |
| DVT | 74/949 | 55/949 | 1.35 (0.96–1.89) | 2 (−0.3–4.3) | 0.083 |
| SVT | 58/952 | 40/964 | 1.47 (0.99–2.18) | 1.94 (0–3.9) | 0.054 |
Propensity score matching applied. BE: Bronchiectasis; wPsA: with Pseudomonas aeruginosa; w/oPsA: without Pseudomonas aeruginosa; ARR: adjusted risk ratio; ARD: adjusted risk difference; CV: cardiovascular; TIA: transient ischaemic attack; AF/AFL: atrial fibrillation and flutter; VA: ventricular arrhythmia; HD: heart disease; IHD: ischaemic heart disease; ACD: acute coronary disease; MI: myocardial infarction; ICM: ischaemic cardiomyopathy; HF: heart failure; NICM: non-ischaemic cardiomyopathy; CS: cardiogenic shock; PE: pulmonary embolism; DVT: deep vein thrombosis; SVT: superficial vein thrombosis. #: these numbers are rounded up to 10 to protect patient privacy, therefore the risk assessment is not disclosed as results may have been impacted.
FIGURE 3.
Forest plot showing the risk ratio for a) individual and b) composite cardiovascular (CV) events in patients with bronchiectasis (BE) with Pseudomonas aeruginosa (wPsA) infection (n=993) compared with propensity-score matched patients with BE without PsA (w/oPsA) (n=993). Composite outcomes consisted of cerebrovascular disorders (stroke and transient ischaemic attack), dysrhythmias (atrial fibrillation or flutter, bradycardia and ventricular arrhythmia), inflammatory heart disease (pericarditis or myocarditis), ischaemic heart disease (acute coronary disease, myocardial infarction, ischaemic cardiomyopathy or angina), other cardiac disorders (heart failure, non-ischaemic cardiomyopathy, cardiac arrest or cardiogenic shock), thrombotic disorders (pulmonary embolism, deep vein thrombosis or superficial vein thrombosis), MACE (stroke, myocardial infarction and all-cause mortality) and any CV event mentioned above. Outcomes were followed in both cohorts for 5 years. Adjusted risk ratios and 95% CIs are presented.
TABLE 4.
Risk of composite cardiovascular events in patients with BE wPsA (n=993) compared with patients with BE w/oPsA (n=993)
| Outcomes | BE wPSA | BE w/oPsA | ARR (95% CI) | ARD % (95% CI) | p-value |
|---|---|---|---|---|---|
| CV accidents | 104/897 | 92/902 | 1.14 (0.87–1.48) | 1.4 (−1.5–4.3) | 0.343 |
| Dysrhythmias | 252/693 | 199/723 | 1.32 (1.13–1.54) | 8.8 (4–13.7) | <0.001 |
| Inflammatory HD | 50/967 | 24/969 | 2.09 (1.29–3.37) | 2.7 (1–4.4) | 0.002 |
| IHD | 93/929 | 87/946 | 1.09 (0.82–1.44) | 0.8 (−1.9–3.5) | 0.550 |
| Other cardiac disorders | 177/829 | 127/831 | 1.40 (1.14–1.72) | 6.1 (2.4–9.8) | 0.001 |
| Thrombotic disorders | 120/884 | 94/905 | 1.31 (1.01–1.68) | 3.2 (0.2–6.2) | 0.038 |
| MACE | 384/869 | 291/886 | 1.35 (1.19–1.52) | 11.3 (6.8–15.9) | < 0.0001 |
| Any CV event | 295/564 | 220/596 | 1.42 (1.24–1.62) | 15.4 (9.7–21.1) | <0.0001 |
Propensity score matching applied. BE: bronchiectasis; wPsA: with Pseudomonas aeruginosa; w/oPsA: without Pseudomonas aeruginosa; ARR: adjusted risk ratio; ARD: adjusted risk difference; CV: cardiovascular; HD: heart disease; IHD: ischaemic heart disease; MACE: major cardiovascular event (defined as non-fatal stroke, non-fatal myocardial infarction and all-cause mortality).
Sensitivity analysis
We conducted additional analyses to validate our findings. Inclusion of laboratory data did not fundamentally change the results of the primary analysis with the exception of a few measures (supplementary tables S8 and S9). Unlike in the primary analysis, the wPsA group was significantly associated with superficial vein thrombosis (RR: 1.84; 95% CI: 1.20–2.82) but not with cardiac arrest (RR: 1.38; 95% CI: 0.79–2.40) or any thrombotic disorders (RR: 1.25; 95% CI: 0.97–1.60). This analysis suggests that these laboratory data minimally affected our primary results. Furthermore, E-values confirmed that there was insignificant influence from unmeasured confounders between PsA and CV events (supplementary table S10).
Discussion
This was a large, retrospective, US population-based cohort study of patients with BE with versus without airway PsA infection with no prior history of CV events before BE diagnosis. It included a longitudinal analysis of CV events occurring over the 5 years following the index BE exacerbation. Patients wPsA airway infection were at greater risk of CV events than PSM patients w/oPsA. Notably, PsA infection was associated with a higher risk of dysrhythmias (including atrial fibrillation and flutter, tachycardia, bradycardia and ventricular arrythmias), inflammatory heart disease (mainly pericarditis), heart failure, cardiac arrest, thrombotic disorders, MACE (although the latter was driven by an increased risk of all-cause mortality) and any CV event as part of a pre-defined composite end-point. However, PsA airway infection was neither associated with MACE components other than all-cause mortality (i.e. non-fatal stroke or myocardial infarction) nor with transient ischaemic attacks. There was also an increased rate of hospitalisation in patients wPsA versus patients w/oPsA. Importantly, this analysis showed that even after adjustment for baseline potential confounders, including demographic factors, comorbidities, pre-existing CV disease, tobacco use, cardiac and BE medications including antibiotics (such as macrolides and quinolones associated with adverse cardiac effects) [19], the risk for CV complications remained higher in the wPsA group after the index exacerbation.
CV sequelae post exacerbation have been described previously in patients with BE in other parts of the world. A UK-based population study showed an increased rate of either first myocardial infarction or stroke in the 91-day period after a respiratory tract infection in patients with BE [9]. A retrospective analysis of a Spanish prospective study demonstrated that ∼30% of patients experienced at least one CV event after an exacerbation during a 35-month follow-up period [7]. However, these studies neither evaluated whether there was a link between the increased risk of CV comorbidities and airway PsA infection nor did they include a comprehensive assessment of CV events following the exacerbation. To the best of our knowledge, our study is the largest and has the longest follow-up period for the assessment of CV outcomes in patients with BE following an exacerbation. Our data suggest for the first time that airway PsA infection directly contributes to the increased CVD risk in patients with BE.
Although our study did not investigate the mechanisms that contribute to the increased CV risk following exacerbations in patients wPsA, it is likely that systemic inflammation, which is detected in patients with BE, is a contributory factor as systemic inflammation is known to promote atherogenesis and trigger acute CV events by inducing endothelial injury/dysfunction (ED), increasing endothelial permeability to lipoproteins and their subendothelial accumulation, promoting leukocyte recruitment in vessel walls, and inducing platelet activation, thrombus formation and plaque rupture [20, 21]. Circulating levels of fibrinogen and CRP are independent risk factors for CVD [22] and are elevated in clinically stable patients with BE [23, 24]. In addition, during an exacerbation, blood levels of these biomarkers increase further [14, 24], and high fibrinogen and CRP levels, driven by chronic PsA infection, have been proposed as biomarkers of disease severity and future exacerbations [25, 26].
Infection wPsA may contribute to the increased systemic inflammation and ED in patients with BE as elevated airway bacterial loads (including PsA) [27, 28] in patients with BE are associated with increased circulating levels of pro-inflammatory mediators. Additional studies are required to investigate the potential mechanistic link between airway PsA infection, systemic inflammation, ED and CV events in patients with BE.
The cause of the increased systemic inflammation and ED in patients with BE who are infected wPsA is unclear. However, this may be caused by spillover of inflammation in the lung into the circulation. Also, lipopolysaccharide (LPS/endotoxin) is released by gram-negative bacteria including PsA, and chronic low-grade endotoxaemia has been strongly linked to systemic inflammation and atherosclerosis [29]. LPS is detected in plasma samples from clinically stable patients with BE (unlike healthy controls), and LPS levels are higher in patients with purulent sputum (who likely have a higher sputum PsA and/or myeloperoxidase burden) than those with mucoid or mucopurulent sputum [30]. It is possible that in patients with BE infected with PsA, circulating levels of PsA-derived LPS increase as the airway PsA burden increases leading to activation of circulating immune cells, causing systemic inflammation, ED, progression of atherosclerosis and/or increased risk of acute CV events.
Pre-existing subclinical CV abnormalities in patients with BE wPsA infection may have contributed to their increased CV risk. Flow-mediated vasodilation (FMD), a measure of ED, is lower in patients with BE than control subjects, and PsA infection is an independent factor for impaired FMD in patients with BE after adjustment for other clinical variables [31]. PsA infection is also an independent risk factor for increased arterial stiffness in patients with BE [32], and arterial stiffness is a strong predictor of CVD [33] by increasing left ventricular afterload and impairing coronary perfusion [34].
Frequent exacerbations may also increase the risk of CV events. Consistent with previous findings [11], our study shows that PsA infection is associated with increased risk of subsequent exacerbations. Patients with BE who are frequent exacerbators (≥3 exacerbations/year) have increased risk factors for CV events including greater arterial stiffness and elevated CRP and troponin levels than infrequent exacerbators (<3 exacerbations/year) [35]. Mechanistically, exacerbations may increase the risk of subsequent CV events by further increasing endotoxaemia, systemic inflammation and ED in patients with BE.
We report for the first time that airway PsA infection was associated with increased risk of any dysrhythmias post index exacerbation. This could potentially be due to the sustained increases in plasma levels of inflammatory cytokines (including tumour necrosis factor-α, interleukin (IL)-1, IL-17 and IL-6) measured in patients with BE who have PsA isolated in their sputum during an exacerbation [14], as these mediators have both direct/indirect arrhythmogenic effects [36, 37]. Thus, PsA-induced inflammation may represent a risk factor for dysrhythmias. Some medications used to treat PsA infection are themselves pro-arrhythmic and should be used with caution in patients at risk of dysrhythmias.
A strength of this study is the large database used to evaluate CV outcomes in patients with BE with or without PsA airway infection. The adjustment for baseline and potential confounders based on PSM strengthened our analyses. Importantly, the algorithm used to identify exacerbations likely reflects clinical practice across healthcare organisations including in primary care, thus increasing generalisability of our exacerbation data. Nevertheless, our study also has several limitations. First, radiological confirmation of the BE diagnosis was not available and our findings may not be generalisable to the entire BE population. Beyond the inherent risk of miscoding in EMRs, a retrospective study recently reported that ICD codes may be insufficient to identify patients with a radiologist diagnosis of BE [38]. A limitation of this study is that PsA-related ICD-10 codes cannot distinguish between PsA colonisation (i.e. chronic infection) and non-chronic infection. It is likely that some patients wPsA infection identified via J15.1 code were underdiagnosed as they were insufficiently symptomatic to have been hospitalised for their PsA infection, which would affect the generalisability of our findings. The lack of routine sputum culture in primary care facilities that contribute to the TriNetX LIVE database may further add to the underdiagnosis of patients wPsA. Unmeasured confounders such as other features of disease severity, economic status, regional factors or different PsA strains may influence outcomes but cannot be controlled for using our PSM approach. Also, our study was neither sufficiently powered nor of sufficient duration to assess the impact of PsA infection on CV mortality. Finally, the likelihood of indirectly identifying CV events may be greater in patients wPsA versus those w/oPsA as the former are more frequently hospitalised than the latter. Future studies with varying follow-up periods may be required to more precisely identify the timing of CV events occurring after an exacerbation in these patients. Future studies with radiographic confirmation of BE, assessment of BE severity with lung function data and symptom score, and confirmation of PsA airway infection with PCR or bacterial culture methods should be conducted to confirm our findings. Determining the cause of hospitalisation and mortality will also improve the interpretation of our findings.
Conclusion
This real-world retrospective analysis of the TriNetX LIVE database supported our hypothesis that PsA infection is associated with poor clinical outcomes in BE including an increased risk of adverse CV events, and all-cause hospitalisations and mortality. The increased risk of CV events post-exacerbation among patients with BE wPsA airway infection encompassed a broad range of CVD including any dysrhythmias, pericarditis, heart failure and cardiac arrest. While PsA infection increased the risk of MACE, this was driven by an increased risk of the all-cause mortality component. Overall, these findings support the hypothesis that airway PsA infection in patients with BE increases the risk of CV events. Current BE treatment goals include improving symptoms and quality of life, preservation of lung function, suppression of bacterial infection and prevention of exacerbations [39]. In addition, PsA eradication therapy is conditionally recommended by the European Respiratory Society Guidelines. The potential effects of antimicrobials on CVD risk and CV event outcomes in patients wPsA who achieved successful eradication remains to be investigated. Similar to COPD, increased awareness of the risk of CV events as part of an integrated care approach may help reduce the CV disease burden in patients with BE [40].
Acknowledgements
The authors thank Gabriel Abreu (Biometrics, Early R&I, BioPharmaceuticals R&D, Gothenburg, Sweden) for plotting the Kaplan–Meier graphical curves and Charles Sopp (Evidence, R&I, Biopharmaceuticals Medical, Gothenburg, Sweden) for checking the quality of the data.
Footnotes
Provenance: Submitted article, peer reviewed.
Author contributions: Conception, design, data analysis and interpretation, and manuscript preparation and review: E. L, C.A. Owen, F. Reid and R. Hughes; manuscript review: C. Cabrera and C.S. Haworth. All authors read and approved the final manuscript.
Conflict of interest: E. L., C.A. Owen, F. Reid, C. Cabrera and R. Hughes are employed by AstraZeneca and may own stock. C.S. Haworth has performed consultancy work/given educational talks for AstraZeneca, BiomX, Chiesi, Infex, Insmed, LifeArc, Pneumagen, Vertex, Zambon and 30 Technology.
Support statement: Funding was provided by AstraZeneca.
Supplementary material
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material
01126-2024.SUPPLEMENT
Data availability
The data that support the findings of this study are available from TriNetX Dataworks Network. For more information, visit https://trinetx.com.
References
- 1.Aliberti S, Goeminne PC, O'Donnell AE, et al. Criteria and definitions for the radiological and clinical diagnosis of bronchiectasis in adults for use in clinical trials: international consensus recommendations. Lancet Respir Med 2022; 10: 298–306. doi: 10.1016/S2213-2600(21)00277-0 [DOI] [PubMed] [Google Scholar]
- 2.McDonnell MJ, Aliberti S, Goeminne PC, et al. Comorbidities and the risk of mortality in patients with bronchiectasis: an international multicentre cohort study. Lancet Respir Med 2016; 4: 969–979. doi: 10.1016/S2213-2600(16)30320-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chalmers JD, Polverino E, Crichton ML, et al. Bronchiectasis in Europe: data on disease characteristics from the European Bronchiectasis registry (EMBARC). Lancet Respir Med 2023; 11: 637–649. doi: 10.1016/S2213-2600(23)00093-0 [DOI] [PubMed] [Google Scholar]
- 4.Marsland I, Sobala R, De Soyza A, et al. Multimorbidity in bronchiectasis: a systematic scoping review. ERJ Open Res 2023; 9: 00296-2022. doi: 10.1183/23120541.00296-2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Gale NS, Bolton CE, Duckers JM, et al. Systemic comorbidities in bronchiectasis. Chron Respir Dis 2012; 9: 231–238. doi: 10.1177/1479972312459973 [DOI] [PubMed] [Google Scholar]
- 6.Navaratnam V, Millett ER, Hurst JR, et al. Bronchiectasis and the risk of cardiovascular disease: a population-based study. Thorax 2017; 72: 161–166. doi: 10.1136/thoraxjnl-2015-208188 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Mendez R, Feced L, Alcaraz-Serrano V, et al. Cardiovascular events during and after bronchiectasis exacerbations and long-term mortality. Chest 2022; 161: 629–636. doi: 10.1016/j.chest.2021.10.013 [DOI] [PubMed] [Google Scholar]
- 8.Sipila PN, Lindbohm JV, Batty GD, et al. Severe infection and risk of cardiovascular disease: a multicohort study. Circulation 2023; 147: 1582–1593. doi: 10.1161/CIRCULATIONAHA.122.061183 [DOI] [PubMed] [Google Scholar]
- 9.Navaratnam V, Root AA, Douglas I, et al. Cardiovascular outcomes after a respiratory tract infection among adults with non-cystic fibrosis bronchiectasis: a general population-based study. Ann Am Thorac Soc 2018; 15: 315–321. doi: 10.1513/AnnalsATS.201706-488OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Choate R, Aksamit TR, Mannino D, et al. Pseudomonas aeruginosa associated with severity of non-cystic fibrosis bronchiectasis measured by the modified bronchiectasis severity score (BSI) and the FACED: The US bronchiectasis and NTM Research Registry (BRR) study. Respir Med 2021; 177: 106285. doi: 10.1016/j.rmed.2020.106285 [DOI] [PubMed] [Google Scholar]
- 11.Araujo D, Shteinberg M, Aliberti S, et al. The independent contribution of Pseudomonas aeruginosa infection to long-term clinical outcomes in bronchiectasis. Eur Respir J 2018; 51: 1701953. doi: 10.1183/13993003.01953-2017 [DOI] [PubMed] [Google Scholar]
- 12.Wilson R, Aksamit T, Aliberti S, et al. Challenges in managing Pseudomonas aeruginosa in non-cystic fibrosis bronchiectasis. Respir Med 2016; 117: 179–189. doi: 10.1016/j.rmed.2016.06.007 [DOI] [PubMed] [Google Scholar]
- 13.Libby P. Inflammation in atherosclerosis: no longer a theory. Clin Chem 2021; 67: 131–142. doi: 10.1093/clinchem/hvaa275 [DOI] [PubMed] [Google Scholar]
- 14.Menendez R, Mendez R, Amara-Elori I, et al. Systemic inflammation during and after bronchiectasis exacerbations: impact of Pseudomonas aeruginosa. J Clin Med 2020; 9: 2631. doi: 10.3390/jcm9082631 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Henkle E, Chan B, Curtis JR, et al. Characteristics and health-care utilization history of patients with bronchiectasis in US medicare enrollees with prescription drug plans, 2006 to 2014. Chest 2018; 154: 1311–1320. doi: 10.1016/j.chest.2018.07.014 [DOI] [PubMed] [Google Scholar]
- 16.Astrand A, Kiddle SJ, Siva Ganesh Mudedla R, et al. Effect of COVID-19 on bronchiectasis exacerbation rates: a retrospective U.S. insurance claims study. Ann Am Thorac Soc 2024; 21: 261–270. doi: 10.1513/AnnalsATS.202211-944OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Blanchette CM, Noone JM, Stone G, et al. Healthcare cost and utilization before and after diagnosis of Pseudomonas aeruginosa among patients with non-cystic fibrosis bronchiectasis in the U.S. Med Sci (Basel) 2017; 5: 20. doi: 10.3390/medsci5040020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Xie Y, Xu E, Bowe B, et al. Long-term cardiovascular outcomes of COVID-19. Nat Med 2022; 28: 583–590. doi: 10.1038/s41591-022-01689-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Mosholder AD, Mathew J, Alexander JJ, et al. Cardiovascular risks with azithromycin and other antibacterial drugs. N Engl J Med 2013; 368: 1665–1668. doi: 10.1056/NEJMp1302726 [DOI] [PubMed] [Google Scholar]
- 20.Gimbrone MA,Jr., Garcia-Cardena G. Endothelial cell dysfunction and the pathobiology of atherosclerosis. Circ Res 2016; 118: 620–636. doi: 10.1161/CIRCRESAHA.115.306301 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Bazaz R, Marriott HM, Francis SE, et al. Mechanistic links between acute respiratory tract infections and acute coronary syndromes. J Infect 2013; 66: 1–17. doi: 10.1016/j.jinf.2012.09.009 [DOI] [PubMed] [Google Scholar]
- 22.Emerging Risk Factors Collaboration , Kaptoge S, Di Angelantonio E, Pennells L, et al. C-reactive protein, fibrinogen, and cardiovascular disease prediction. N Engl J Med 2012; 367: 1310–1320. doi: 10.1056/NEJMoa1107477 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Loebinger MR, Shoemark A, Berry M, et al. Procalcitonin in stable and unstable patients with bronchiectasis. Chron Respir Dis 2008; 5: 155–160. doi: 10.1177/1479972308088823 [DOI] [PubMed] [Google Scholar]
- 24.Brill SE, Patel AR, Singh R, et al. Lung function, symptoms and inflammation during exacerbations of non-cystic fibrosis bronchiectasis: a prospective observational cohort study. Respir Res 2015; 16: 16. doi: 10.1186/s12931-015-0167-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Posadas T, Oscullo G, Zaldivar E, et al. C-reactive protein concentration in steady-state bronchiectasis: prognostic value of future severe exacerbations. Data from the Spanish Registry of Bronchiectasis (RIBRON). Arch Bronconeumol (Engl Ed) 2021; 57: 21–27. doi: 10.1016/j.arbr.2019.12.022 [DOI] [PubMed] [Google Scholar]
- 26.Lee SJ, Jeong JH, Heo M, et al. Serum fibrinogen as a biomarker for disease severity and exacerbation in patients with non-cystic fibrosis bronchiectasis. J Clin Med 2022; 11: 3948. doi: 10.3390/jcm11143948 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Sibila O, Laserna E, Shoemark A, et al. Airway bacterial load and inhaled antibiotic response in bronchiectasis. Am J Respir Crit Care Med 2019; 200: 33–41. doi: 10.1164/rccm.201809-1651OC [DOI] [PubMed] [Google Scholar]
- 28.Chalmers JD, Smith MP, McHugh BJ, et al. Short- and long-term antibiotic treatment reduces airway and systemic inflammation in non-cystic fibrosis bronchiectasis. Am J Respir Crit Care Med 2012; 186: 657–665. doi: 10.1164/rccm.201203-0487OC [DOI] [PubMed] [Google Scholar]
- 29.Mohammad S, Thiemermann C. Role of metabolic endotoxemia in systemic inflammation and potential interventions. Front Immunol 2020; 11: 594150. doi: 10.3389/fimmu.2020.594150 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Owen CA, Campbell EJ, Hill SL, et al. Increased adherence of monocytes to fibronectin in bronchiectasis. Regulatory effects of bacterial lipopolysaccharide and role of CD11/CD18 integrins. Am Rev Respir Dis 1992; 145: 626–631. doi: 10.1164/ajrccm/145.3.626 [DOI] [PubMed] [Google Scholar]
- 31.Gao YH, Liu SX, Cui JJ, et al. Subclinical atherosclerosis in adults with steady-state bronchiectasis: a case-control study. Respir Med 2018; 134: 110–116. doi: 10.1016/j.rmed.2017.11.024 [DOI] [PubMed] [Google Scholar]
- 32.Gao YH, Cui JJ, Wang LY, et al. Arterial stiffness in adults with steady-state bronchiectasis: association with clinical indices and disease severity. Respir Res 2018; 19: 86. doi: 10.1186/s12931-018-0790-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Vlachopoulos C, Aznaouridis K, Stefanadis C. Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and meta-analysis. J Am Coll Cardiol 2010; 55: 1318–1327. doi: 10.1016/j.jacc.2009.10.061 [DOI] [PubMed] [Google Scholar]
- 34.Davies JI, Struthers AD. Pulse wave analysis and pulse wave velocity: a critical review of their strengths and weaknesses. J Hypertens 2003; 21: 463–472. doi: 10.1097/00004872-200303000-00004 [DOI] [PubMed] [Google Scholar]
- 35.Saleh AD, Kwok B, Brown JS, et al. Correlates and assessment of excess cardiovascular risk in bronchiectasis. Eur Respir J 2017; 50: 1701127. doi: 10.1183/13993003.01127-2017 [DOI] [PubMed] [Google Scholar]
- 36.Lazzerini PE, Abbate A, Boutjdir M, et al. Fir(e)ing the rhythm: inflammatory cytokines and cardiac arrhythmias. JACC Basic Transl Sci 2023; 8: 728–750. doi: 10.1016/j.jacbts.2022.12.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Lazzerini PE, Laghi-Pasini F, Boutjdir M, et al. Inflammatory cytokines and cardiac arrhythmias: the lesson from COVID-19. Nat Rev Immunol 2022; 22: 270–272. doi: 10.1038/s41577-022-00714-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Green O, Liautaud S, Knee A, et al. Measuring accuracy of International Classification of Diseases codes in identification of patients with non-cystic fibrosis bronchiectasis. ERJ Open Res 2024; 10: 00715-2023. doi: 10.1183/23120541.00715-2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Polverino E, Goeminne PC, McDonnell MJ, et al. European Respiratory Society guidelines for the management of adult bronchiectasis. Eur Respir J 2017; 50: 1700629. doi: 10.1183/13993003.00629-2017 [DOI] [PubMed] [Google Scholar]
- 40.Hurst JR, Gale CP. Global Working Group on Cardiopulmonary Risk . MACE in COPD: addressing cardiopulmonary risk. Lancet Respir Med 2024; 12: 345–348. doi: 10.1016/S2213-2600(24)00038-9 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material
01126-2024.SUPPLEMENT
Data Availability Statement
The data that support the findings of this study are available from TriNetX Dataworks Network. For more information, visit https://trinetx.com.