The effect of tiotropium/olodaterol versus fluticasone propionate/salmeterol on left ventricular filling and lung hyperinflation in patients with COPD.

This exploratory, randomised, double-blind, double-dummy, multicentre, cross-over study explored the effect of 6 weeks of treatment with tiotropium/olodaterol (T/O) versus fluticasone propionate/salmeterol (F/S) on left ventricular filling in patients with chronic obstructive pulmonary disease with functional residual capacity (FRC) >120% predicted and postbronchodilator improvement of FRC ≥7.5%. Overall, 76 patients were randomised across nine sites. Treatment with T/O or F/S increased left ventricular end-diastolic volume index from baseline (adjusted mean change: T/O: 2.317 mL/m2, F/S: 2.855 mL/m2), with no statistically significant difference between treatments. However, T/O resulted in a significantly greater reduction in lung hyperinflation versus F/S (FRC plethysmography absolute change from baseline: F/S: -0.329 L, T/O: -0.581 L). © Author(s) (or their employer(s)) 2020. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

RCT Entities:   Population Interventions Outcomes

Introduction

In patients with chronic obstructive pulmonary disease (COPD), cardiovascular disease is serious and prevalent.1 2 Several processes may link COPD and cardiovascular disease,3 notably left ventricular (LV) diastolic dysfunction with reduced filling of the left ventricle.4 5 It is proposed that lung hyperinflation results in reduced LV filling,5 which is, in turn, associated with reduced exercise capacity and physical activity.4 5

The combination of the long-acting muscarinic antagonist (LAMA)/long-acting β2-agonist (LABA) tiotropium/olodaterol (T/O) improves lung hyperinflation and exercise tolerance vs comparators.6 Other combinations of LAMA/LABA and LABA/inhaled corticosteroid (ICS) have been shown to reduce lung hyperinflation and improve LV function versus placebo.7 8

We explored the effect of T/O Respimat versus fluticasone propionate/salmeterol (F/S) Accuhaler on LV filling, assessed by cardiovascular MRI (CMRI). Secondary outcomes included lung hyperinflation and other measures of cardiac function.

Methods

Study design

This was an exploratory, randomised, double-blind, double-dummy, multicentre, cross-over study (NCT03055988) evaluating the effect of treatment with T/O 5/5 µg versus F/S 1000/100 µg in patients with COPD. The trial consisted of a run-in period, two 6-week treatment periods with no washout and a follow-up period. During each treatment period, the patient inhaled two puffs from the Respimat inhaler and one inhalation from the Accuhaler in the morning, and one inhalation from the Accuhaler only in the evening; the Respimat inhaler was not used in the evening.

Inclusion/exclusion criteria

Patients aged ≥40–75 years with a smoking history of >10 pack-years were eligible for inclusion if they had postbronchodilator (400 µg salbutamol) forced expiratory volume in 1 s (FEV1) <70% predicted normal, with postbronchodilator FEV1/forced vital capacity (FVC) <70%. Furthermore, they had to have had a prebronchodilator functional residual capacity (FRC) >120% predicted with a postbronchodilator reversibility of at least 7.5% at screening, in line with the criteria of Stone et al.8

Patients were excluded if they had a significant disease other than COPD or a current diagnosis of asthma. Patients who experienced exacerbations in the 6 weeks prior to screening were excluded, as were patients who experienced COPD exacerbations or respiratory tract infections before randomisation. In addition, patients were also excluded if they had a history of myocardial infarction, cerebrovascular event or coronary artery intervention other than coronary artery bypass graft within 1 year of screening, or abnormal ECG with an event such as left bundle branch block and LV hypertrophy. Moreover, patients who had been hospitalised for heart failure within the past year, had current severe heart failure class IV or ejection fraction ≤40% from CMRI baseline assessment, or systolic blood pressure >140 mm Hg and/or diastolic blood pressure >90 mm Hg at screening were also excluded. Patients with stable arterial hypertension controlled under therapy and normotensive at screening could enter the study.

Assessments CMRI

Assessments included CMRI at 1.5 or 3 Tesla lasting ≤1 hour. We assessed cardiac functional parameters and structure using a cardiopulmonary acquisition protocol. Cardiac cine images were acquired in two-chamber, four-chamber and short-axis views. Strain and strain rate of the right and left ventricles were assessed. Aortic distensibility and pulmonary artery pulsatility were derived from cine images acquired at end expiration in planes perpendicular to the thoracic aorta at the level of the pulmonary artery, and perpendicular to the main pulmonary artery. Scans were digitally transferred for blinded central review by an experienced independent reviewer (K-FK).

Lung function

Body plethysmography, including measurement of FRC, residual volume (RVol), inspiratory capacity (IC), total lung capacity (TLC) and forced spirometry, including measurement of FEV1 and FVC, were performed at screening, baseline and after 6 weeks of treatment, in line with American Thoracic Society/European Respiratory Society standards.9 Body plethysmography was followed by forced spirometry measurements. FRC was measured at least three times until three FRC values within 5% variability were obtained. Lung function was measured at the baseline visit 1 hour prior to inhalation of morning dose. At week 6, lung function was measured 1.5 hours after the morning dose of medication.

Endpoints

The primary endpoint was change from baseline in LV end-diastolic volume index (LVEDVI) after 6 weeks of treatment with T/O versus F/S. Secondary endpoints were change from baseline at week 6 in aortic distensibility, pulmonary artery pulsatility, FRC, FEV1 and FVC. Further endpoints assessed after 6 weeks of treatment included TLC, RVol and IC.

Statistical analysis

The full analysis set (FAS) included all randomised patients taking any dose of trial medication, and with both baseline and any evaluable postbaseline measurement for primary or secondary endpoints. Presented results are from FAS patients who did not have an exacerbation during treatment. A restricted maximum likelihood-based mixed-effect repeated measures model was used for analysis of primary and secondary endpoints. This included treatment and period as fixed effects, patient as a random effect and baseline as covariate. All calculated p values for analyses of secondary endpoints are descriptive as no adjustment for multiple testing was carried out.

Results

Overall, 76 patients were randomised and treated across nine centres; 67 (88.2%) completed all treatment periods. Most patients were male (59.2%); mean age was 62 years. All patients were ex-smokers (43.4%) or current smokers (56.6%). Patients had mean FRC% predicted of 146.8% and postbronchodilator FEV1% predicted normal of 52.9%. Clinically stable comorbidities included hypertension (61.8%), coronary artery disease (9.2%), diabetes mellitus (2.6%) and chronic cardiac failure (2.6%) (table 1).

Table 1

Demographics and baseline patient characteristics of treated set

Characteristic	Total (N=76)
Male, n (%)	45 (59.2)
Age, mean (SD), years	61.9 (7.1)
Smoking history, n (%)
Current	43 (56.6)
Former	33 (43.4)
BMI, mean (SD), kg/m2	26 (5.6)
Postbronchodilator % predicted normal FEV1 (SD)	52.9 (12.1)
Postbronchodilator % predicted normal FRC (SD)	146.8 (28.5)
Difference between predose and postdose FRC % predicted (SD)	–18.3 (7.3)
GOLD, n (%)
I (≥80%)	0 (0.0)
II (50–<80%)	46 (60.5)
III (30–<50%)	26 (34.2)
IV (<30%)	4 (5.3)
Cardiac disorders	15 (19.7)
Chronic cardiac failure	2 (2.6)
Coronary artery disease	7 (9.2)
Atrial fibrillation	1 (1.3)
Hypertension	47 (61.8)
Diabetes mellitus	2 (2.6)
No of subjects with at least one pulmonary medication at baseline	71 (93.4)
LAMA	53 (69.7)
SAMA	3 (3.9)
LABA	59 (77.6)
SABA	43 (56.6)
Mucolytics	2 (2.6)
ICS	13 (17.1)
Steroids (oral)	1 (1.3)

BMI, body mass index; FEV1, forced expiratory volume in 1 s; FRC, functional residual capacity; GOLD, Global Initiative for Chronic Obstructive Lung Disease; ICS, inhaled corticosteroid; LABA, long-acting β2-agonist; LAMA, long-acting muscarinic antagonist; SABA, short-acting β2-agonist; SAMA, short-acting muscarinic antagonist.

At week 6, mean LVEDVI was increased from baseline (66.644 mL/m2) by 2.317 mL/m2 (95% CI 0.061 to 4.574) for T/O and 2.855 mL/m2 (95% CI 0.597 to 5.112) for F/S. There was no significant difference between T/O and F/S (treatment difference (TD) –0.537 mL/m2; 95% CI –2.779 to 1.705; p=0.6331) (figure 1A). Neither treatment produced a significant change from baseline in aortic distensibility or pulmonary artery pulsatility. There were no significant TDs in other measures of cardiac function: LV or right ventricular (RV) end-systolic volume index, LV or RV ejection fraction, LV cardiac index, LV or RV stroke volume index, or LV or RV mass index (table 2).

Figure 1

Effect of (A) T/O versus F/S on LVEDVI at week 6 and (B) T/O versus F/S on FRC plethysmograph (percent predicted and absolute change) at week 6. *P<0.01. F/S, fluticasone propionate/salmeterol; FRC, functional residual capacity; LVEDVI, left ventricular end-diastolic volume index; T/O, tiotropium/olodaterol.

Table 2

Cardiac function parameters and arterial stiffness at week 6

	Treatment (n)	Adjusted mean (SE)	Change from baseline(95% CI)	Treatment difference(95% CI)	P value
Cardiac function
LVEDVI (mL/m2)	T/O (59)	68.711 (1.136)	2.317 (0.061 to 4.574)	–0.537 (–2.779 to 1.705)	0.6331
	F/S (59)	69.249 (1.137)	2.855 (0.597 to 5.112)
RVEDVI (mL/m²)	T/O (59)	76.237 (1.334)	2.448 (–0.207 to 5.103)	–0.864 (–3.069 to 1.340)	0.4356
	F/S (59)	77.102 (1.335)	3.312 (0.656 to 5.968)
LVESVI (mL/m2)	T/O (59)	27.324 (0.804)	0.647 (–0.951 to 2.245)	–0.340 (–1.841 to 1.161)	0.6516
	F/S (59)	27.665 (0.804)	0.987 (–0.612 to 2.585)
RVESVI (mL/m2)	T/O (59)	35.912 (1.074)	0.236 (–1.894 to 2.366)	0.257 (–2.152 to 2.666)	0.8315
	F/S (59)	35.655 (1.074)	–0.021 (–2.152 to 2.110)
CARDIDX (L/min/m2)	T/O (59)	3.044 (0.075)	0.044 (–0.104 to 0.193)	–0.036 (–0.201 to 0.130)	0.6687
	F/S (59)	3.080 (0.075)	0.080 (–0.068 to 0.228)
LVSVI (mL/m2)	T/O (59)	41.097 (0.950)	1.523 (–0.362 to 3.409)	–0.628 (–2.676 to 1.421)	0.5240
	F/S (59)	41.725 (0.951)	2.151 (0.264 to 4.038)
RVSVI (mL/m2)	T/O (59)	39.872 (1.046)	1.768 (–0.307 to 3.842)	–1.685 (–4.039 to 0.669)	0.1572
	F/S (59)	41.557 (1.046)	3.453 (1.377 to 5.529)
LVEF (%)	T/O (59)	60.936 (0.958)	0.507 (–1.397 to 2.410)	0.139 (–1.770 to 2.049)	0.8845
	F/S (59)	60.797 (0.959)	0.368 (–1.537 to 2.272)
RVEF (%)	T/O (59)	53.103 (1.124)	0.824 (–1.404 to 3.052)	–0.833 (–3.700 to 2.035)	0.5633
	F/S (59)	53.936 (1.125)	1.657 (–0.572 to 3.886)
LVMI (g/m2)	T/O (59)	50.129 (1.000)	–0.125 (–2.112 to 1.861)	1.021 (–0.968 to 3.010)	0.3083
	F/S (59)	49.108 (1.001)	–1.146 (–3.134 to 0.841)
RVMI (g/m2)	T/O (59)	18.106 (0.605)	0.092 (–1.107 to 1.292)	0.754 (–0.737 to 2.244)	0.3155
	F/S (59)	17.352 (0.606)	–0.661 (–1.862 to 0.539)
Central systolic blood pressure (mm Hg)	T/O (58)	115.395 (1.559)	2.271 (–0.820 to 5.363)	2.069 (–1.640 to 5.779)	0.2687
	F/S (58)	113.325 (1.559)	0.202 (–2.891 to 3.294)
Pulse pressure (mm Hg)	T/O (58)	45.025 (1.014)	0.579 (–1.430 to 2.588)	0.409 (–2.264 to 3.082)	0.7604
	F/S (58)	44.616 (1.014)	0.170 (–1.839 to 2.179)
Arterial stiffness
Aortic distensibility(%/mm Hg)	T/O (57)	0.514 (0.036)	–0.005 (–0.076 to 0.066)	0.001 (–0.072 to 0.074)	0.9817
	F/S (57)	0.513 (0.036)	–0.006 (–0.077 to 0.065)
PA pulsatility (%)	T/O (57)	30.053 (1.836)	1.105 (–2.540 to 4.751)	1.280 (–2.719 to 5.279)	0.5238
	F/S (57)	28.773 (1.837)	–0.175 (–3.823 to 3.474)

CARDIDX, cardiac index; F/S, fluticasone propionate/salmeterol; LVEDVI, left ventricular end-diastolic volume index; LVEF, left ventricular ejection fraction; LVESVI, left ventricular end-systolic volume index; LVMI, left ventricular mass index; LVSVI, left ventricular stroke volume index; PA, pulmonary artery; RVEDVI, right ventricular end-diastolic volume index; RVEF, right ventricular ejection fraction; RVESVI, right ventricular end-systolic volume index; RVMI, right ventricular mass index; RVSVI, right ventricular stroke volume index; T/O, tiotropium/olodaterol.

FRC% predicted was reduced from baseline by 10.211% with F/S and 18.168% with T/O. There was a significant difference in favour of T/O, with a between-group difference of –7.957% (95% CI –12.865 to –3.050; p=0.0019) (figure 1B). Absolute change from baseline in FRC (plethysmograph) was –0.329 L with F/S and –0.581 L with T/O, with a between-group difference of –0.252 L in favour of T/O (95% CI –0.413 to –0.091; p=0.0028) (figure 1B). There was also a larger reduction in RVol with T/O (–0.572 L) than with F/S (–0.321 L; TD –0.251 L; 95% CI –0.409 to –0.093; p=0.0024). Both drugs increased IC compared with baseline (0.320 L and 0.289 L for T/O and F/S, respectively) with no significant difference between treatments. There was a greater reduction in TLC with T/O (–0.206 L) than with F/S (–0.023 L; TD between T/O and F/S –0.184 L; 95% CI –0.358 to –0.010; p=0.0390) (table 3).

Table 3

Hyperinflation and spirometry at week 6

	Treatment (n)	Adjusted mean (SE)	Change from baseline(95% CI)	Treatment difference(95% CI)	P value
Lung hyperinflation
FRC plethysmography (%)	T/O (59)	146.783 (2.065)	–18.168 (–22.262 to –14.074)	–7.957 (–12.865 to –3.050)	0.0019
	F/S (59)	154.740 (2.066)	–10.211 (–14.308 to –6.113)
FRC (L)	T/O (59)	4.770 (0.069)	–0.581 (–0.718 to –0.443)	–0.252 (–0.413 to –0.091)	0.0028
	F/S (59)	5.022 (0.069)	–0.329 (–0.466 to –0.191)
RVol (L)	T/O (58)	3.742 (0.067)	–0.572 (–0.704 to –0.439)	–0.251 (–0.409 to –0.093)	0.0024
	F/S (58)	3.993 (0.067)	–0.321 (–0.453 to –0.188)
IC (L)	T/O (58)	2.066 (0.061)	0.320 (0.198 to 0.433)	0.032 (–0.084 to 0.148)	0.5832
	F/S (58)	2.035 (0.061)	0.289 (0.166 to 0.411)
TLC (L)	T/O (58)	6.962 (0.071)	–0.206 (–0.347 to –0.066)	–0.184 (–0.358 to –0.010)	0.0390
	F/S (58)	7.146 (0.071)	–0.023 (–0.163 to 0.118)
Spirometry
FEV1 (L)	T/O (59)	1.725 (0.037)	0.339 (0.266 to 0.412)	0.180 (0.121 to 0.240)	<0.0001
	F/S (59)	1.545 (0.037)	0.158 (0.086 to 0.231)
FVC (L)	T/O (59)	3.556 (0.054)	0.445 (0.338 to 0.552)	0.286 (0.171 to 0.400)	<0.0001
	F/S (59)	3.270 (0.054)	0.159 (0.052 to 0.266)

FEV1, forced expiratory volume in 1 s; FRC, functional residual capacity; F/S, fluticasone propionate/salmeterol; FVC, forced vital capacity; IC, inspiratory capacity; RVol, residual lung volume; TLC, total lung capacity; T/O, tiotropium/olodaterol.

T/O was associated with larger improvements from baseline for FEV1 (0.339 L vs 0.158 L with F/S; TD 0.180 L, 95% CI 0.121 to 0.240; p<0.0001), and for FVC 0.445 L vs 0.159 L with F/S; TD 0.286 L, 95% CI 0.171 to 0.400; p<0.0001).

Overall, the safety profile in the study was consistent with the known safety profile of both treatments.

Discussion

To our knowledge, this is the first study to explore the effect of combining two bronchodilators versus an active treatment (LABA/ICS) on cardiac function. Both treatments improved LV function and decreased lung hyperinflation from baseline. Although T/O provided greater improvements in lung hyperinflation than F/S at week 6, there was no significant between-treatment difference in LV or RV filling. While pulmonary function testing was performed with fixed timing 1.5 hours post-treatment, CMRI was conducted 1–8 hours post-treatment, which may partly explain the lack of difference in cardiac outcome compared with pulmonary function.

Two previous monocentric studies found improvements in hyperinflation that translated into increases in LV function, although neither used an active comparator. The CLAIM study showed a decrease in RVol with indacaterol/glycopyrronium, and a significant improvement in cardiac function versus placebo.7 Another study reported improved RVol with fluticasone furoate/vilanterol vs placebo, also translating into increases in RV and LVEDVI due to improved cardiac filling.8

The baseline LVEDVI and comorbid hypertension among patients in these studies was lower than in those in our study. Perhaps the mild severity of the analysed population in our study, with baseline mean LVEDVI of 66.64 mL/m2, plus the higher prevalence of hypertension may have compensated for any between-treatment differences. In patients with COPD and lung hyperinflation, 6 weeks of treatment with T/O or F/S increased LVEDVI from baseline, with no statistically significant difference between treatments. However, T/O resulted in a significantly greater reduction in lung hyperinflation versus F/S. Our finding that LAMA/LABA or LABA/ICS improve hyperinflation and cardiac function from baseline is consistent with previous studies.7 8

Lay summary

Many people with COPD die from heart disease. COPD causes air-trapping in the lungs (lung hyperinflation) and reduced filling of the bottom-left chamber (ventricle) of the heart. T/O is an inhaled medicine that can make the lungs work more effectively, reduce breathlessness and improve quality of life by opening the airways in people with COPD.

We wanted to test if T/O is better than another inhaler—F/S—in improving how the lungs and heart work. We measured the ability of patients to expel air as well as assessing changes in heart function using cardiac imaging techniques after 6 weeks of treatment.

We found that both T/O and F/S improved lung and heart function. Although T/O did improve lung function by reducing lung hyperinflation more than F/S after 6 weeks, both treatments resulted in similar improvements in heart filling.