Tiotropium in chronic obstructive pulmonary disease – a review of clinical development

In a 4-week, dose-response study of tiotropium inhalation powder (4.5, 9, 18, and 36 μg QD) in patients with COPD (n = 169), all doses significantly improved lung function compared with placebo (day 29; difference in trough forced expiratory volume in 1 s [FEV₁] response = 0.14 L, 0.11 L, 0.15 L, and 0.19 L, respectively; p < 0.05 for all) [31]. Similarly, forced vital capacity (FVC) was higher for all doses of tiotropium than with placebo (day 29; difference in trough FVC response = 0.31 L, 0.22 L, 0.37 L, and 0.22 L, respectively; p < 0.05 for 4.5-μg and 18.0-μg doses). All tiotropium doses also increased peak expiratory flow rate (PEFR) compared with placebo at all test points (morning, noon, and evening) [31]. Improvement in lung function was maintained throughout the treatment period. The overall safety profile of the four tiotropium doses was similar to that of placebo [31]. Based on these results, a dosage of 18 μg QD was evaluated in a 3-month, randomized controlled trial (RCT; n = 470) [21]. Again, tiotropium significantly improved lung function compared with placebo (day 92; difference in trough FEV₁ response = 0.15 L; difference in trough FVC response = 0.28 L; p < 0.001 for both), and improvements were maintained throughout the treatment period. Tiotropium also significantly improved daily morning and evening PEFR and reduced symptoms of wheezing, shortness of breath, and “as-needed” albuterol use. Safety of tiotropium was similar to that of placebo, except for dry mouth, which was more common with tiotropium than with placebo (9.3% vs 1.6%; p < 0.05) [21]. Further, the efficacy and safety of tiotropium 18 μg QD was compared with that of ipratropium 40 μg QID (via a pMDI) in a 13-week study (n = 288) [40]. Tiotropium improved lung function compared with ipratropium throughout the treatment period (day 92; mean difference in trough FEV₁ response = 0.13 L [p = 0.0001]; mean difference in trough FVC response = 0.21 L [p = 0.0003]). Tiotropium also consistently improved weekly mean morning and evening PEFR and reduced the use of rescue salbutamol. As observed in other studies, dry mouth was more common with tiotropium than with ipratropium (14.7% vs 10.3%) [40].

In a 6-month RCT (n = 623), tiotropium 18 μg QD compared with the LABA salmeterol 50 μg twice daily (BID; delivered via a pMDI) significantly improved lung function (week 24; difference in trough FEV₁ response = 0.052 L; difference in trough FVC response = 0.112 L; p < 0.01 for both) and reduced dyspnea (week 24; difference in transition dyspnea index [TDI] focal score = 0.78 U; p < 0.05) [35]. Both active drugs significantly reduced the need for rescue albuterol compared with placebo (p < 0.0001). In this study, tiotropium also significantly improved HRQoL compared with placebo (week 24; difference in St. George’s Respiratory Questionnaire [SGRQ] total score = − 2.71 U; p < 0.05) but not salmeterol (difference = − 1.6 U; p > 0.05). Safety of tiotropium was similar to that of placebo and salmeterol, except for dry mouth, which was more common with tiotropium (10%) [35]. Health outcomes were evaluated in two 6-month RCTs (n = 1207) in which tiotropium 18 μg compared with placebo significantly reduced the number of exacerbations/patient-year (PY; 1.07 vs 1.49; p < 0.05), increased the time to first exacerbation (p ≤ 0.01), and improved HRQoL (difference in SGRQ total score = − 2.7 U; p < 0.01) during the treatment period [36]. Salmeterol did not significantly differ from placebo for these outcomes. Although both active drugs improved lung function, tiotropium did so to a greater extent than salmeterol. Dry mouth was more common with tiotropium (8.2%) than with salmeterol (1.7%) or placebo (2.3%) [36].

In two identical 12-month RCTs (n = 921), tiotropium 18 μg QD significantly improved lung function compared with placebo (difference in trough FEV₁ response = 0.12–0.15 L [range over days 1–344]; p < 0.01) [22]. In addition, compared with patients who received placebo, those who received tiotropium also had significantly less dyspnea (difference in TDI focal score = 0.8–1.1 U [range over days 50–344]; p < 0.001), better health status scores (p < 0.05), and fewer COPD exacerbations/PY (0.76 vs 0.95; p = 0.045) and exacerbation-related hospitalizations (p < 0.05). In this and other tiotropium trials, patients were allowed to continue taking glucocorticoids during the study period. As observed in prior studies, incidence of dry mouth was higher with tiotropium than with placebo (16.0% vs 2.7%; p < 0.05). In two other identical 12-month RCTs (n = 535), tiotropium 18 μg QD compared with ipratropium 40 μg QID (delivered via a pMDI) significantly improved lung function (difference in trough FEV₁ response = 0.15 L; p < 0.001) and reduced the use of rescue salbutamol (p < 0.05) at the end of the treatment period [34]. In addition, tiotropium compared with ipratropium significantly reduced dyspnea (day 364; difference in TDI focal score = 0.90 U; p = 0.001) and the number of exacerbations/PY (0.73 vs 0.96; p = 0.006); improved PEFR (difference in morning PEFR = 10–18 L/min [range over days 1–365]; difference in evening PEFR = 9–18 L/min [range over days 1–365]; p < 0.01) and HRQoL (day 364; difference in SGRQ total score = − 3.30; p = 0.004); and increased the time to first exacerbation (p = 0.008) and time to first exacerbation-related hospitalization (p = 0.048). These improvements were maintained throughout the treatment period. Collectively, results of these studies showed that QD tiotropium 18 μg was a safe and efficacious LAMA for maintenance treatment of COPD and for reducing exacerbations, leading to the approval of Spiriva® HandiHaler® in the EU (2002), US (2004), and other countries [41, 42].

Patients with COPD often experience hyperinflation, which results in reduced inspiratory capacity (IC), limited exercise capacity, and increased exertional dyspnea [2, 43]. A few RCTs were designed to specifically evaluate the effects of tiotropium on these parameters.

In a 4-week RCT (n = 81), tiotropium 18 μg QD significantly improved lung function and IC compared with placebo (mean differences at week 4: trough FEV₁ = 0.16 L; trough FVC = 0.33 L; trough IC = 0.22 L; p < 0.01 for all) [44]. Similarly, in a 6-week RCT (n = 187), tiotropium significantly improved lung function compared with placebo, consistent with findings from prior studies [45]. Further, tiotropium compared with placebo significantly reduced lung hyperinflation (residual volume, p < 0.001; functional residual capacity, p < 0.001); increased vital capacity (p < 0.0001), IC (difference in trough response = 0.10 L; p < 0.05), and exercise endurance (difference in endurance time = 105 s [21%]; p < 0.01); and decreased exertional dyspnea (p < 0.01) on day 42 [45]. Change from baseline in all parameters was evident after the first treatment and persisted for the duration of the trial. In a subsequent 6-week study (n = 261), effects of tiotropium on lung hyperinflation, symptom-limited exercise tolerance, and exertional dyspnea were apparent at 2.25 h of treatment and lasted for 8 h after dosing on day 42 [46]. These findings were corroborated by results of another RCT (n = 100), where tiotropium 18 μg compared with placebo not only significantly improved trough FVC (difference = 0.20 L; p < 0.05) and trough IC (difference = 0.15 L; p < 0.05) but also significantly increased mean distance walked during the shuttle-walk test (difference = 36 m; p < 0.05) and improved HRQoL (difference in SGRQ total score = − 6.5; p = 0.026) after 12 weeks of treatment [32]. Further, in an RCT (n = 554) by the Tiotropium: Influence sur la Perception de l’amélioration des activites Habituelles Objectivée par une echelle Numerique (TIPHON) group—in addition to significantly improving lung function (difference in trough FEV₁ = 0.10 L; p = 0.0001) and reducing exacerbations/PY (1.05 vs 1.83; p = 0.0287)—tiotropium compared with placebo significantly increased the proportion of patients achieving clinically relevant improvement in HRQoL (difference in responders = 10.9%; p = 0.029) after 9 months of treatment [23]. In contrast to the results of the above studies, in a 96-week RCT (n = 519), the difference in endurance time between tiotropium 18 μg QD and placebo was not statistically significant (tiotropium/placebo = 1.13; 95% CI = 0.97–1.32; p = 0.106). However, consistent with previous studies, tiotropium improved HRQoL at 96 weeks compared with placebo (difference in SGRQ total score = 4.03 U; p = 0.007) [47].

Patients with COPD, even those with mild disease, can experience exacerbations [48‐51], which account for a large proportion of total COPD burden on the healthcare system [52, 53]. Recommendations for reducing COPD exacerbations and treatment of stable COPD have been provided by GOLD [2], the European Respiratory Society/American Thoracic Society [54], and the Spanish Society of Pulmonology and Thoracic Surgery [55, 56]. The efficacy of tiotropium 18 μg QD in reducing exacerbations was evaluated in several short- and long-term studies, including those mentioned above (Table 1; summarized by Anzueto et al. [57]).

In an RCT (n = 1829) conducted in a US Veterans Affairs setting, fewer patients treated with tiotropium than placebo experienced ≥1 exacerbation (difference = − 5.7%; p = 0.037) or exacerbation-related hospitalization (difference = − 3.0%; p = 0.056) after 6 months of treatment [37]. Analysis of secondary outcomes indicated that tiotropium significantly increased time to first exacerbation (p = 0.028) and reduced health care resource utilization (HCRU; frequency of hospitalizations, p = 0.047; antibiotic treatment days, p = 0.015; and unscheduled clinic visits, p = 0.019). Similar results were observed in another RCT (n = 1010; Mesure de l’Influence de Spiriva® sur les Troubles Respiratoires Aigus à Long terme [MISTRAL]) [24], where tiotropium compared with placebo significantly increased time to first exacerbation (p < 0.001) and reduced the number of exacerbations/PY (1.57 vs 2.41; p < 0.001), proportion of patients with ≥1 exacerbation (difference = − 10.4%; p < 0.01), and HCRU (concomitant respiratory medications, p < 0.0001; antibiotics, p < 0.001; and oral steroids, p < 0.01; and the number of unscheduled physician contacts, p < 0.05) after treatment for 1 year [24]. Tiotropium also significantly improved weekly morning PEFR (mean difference over 1 year = 25 L/min; p < 0.0001), trough FEV₁ (mean difference = 0.12 L; p < 0.0001), FVC (mean difference = 0.17 L; p < 0.0001), and IC (mean difference = 0.14 L; p < 0.001) at the end of the treatment period. As observed in other studies, the safety of tiotropium was similar to that of placebo, except for dry mouth, which was more frequent with tiotropium (4.0%) than with placebo (1.4%) [24]. In addition, in the Prevention Of Exacerbations with Tiotropium in COPD (POET-COPD) trial (n = 7376), tiotropium 18 μg QD was significantly more efficacious than salmeterol 50 μg BID in increasing the time to first exacerbation (hazard ratio [HR] = 0.83 [ie, 17% reduction in risk of exacerbations with tiotropium]; p < 0.001) and reducing the annual rate of moderate or severe exacerbations (0.64 vs 0.72; rate ratio [RR] = 0.89 [11% reduction with tiotropium]; p = 0.002) and severe exacerbations (0.09 vs 0.13; RR = 0.73 [ie, 27% reduction with tiotropium]; p < 0.001) after 1 year of treatment [39]. The safety of tiotropium was similar to that of salmeterol. In a post hoc analysis of POET-COPD, tiotropium compared with salmeterol increased the time to first exacerbation and reduced the number of exacerbations in patients at low and high risk of exacerbation (time to first exacerbation: HR = 0.89; p = 0.1046 and HR = 0.84; p = 0.0002, respectively; number of exacerbations: RR = 0.89; p = 0.1768 and RR = 0.90; p = 0.0383, respectively) [58].

Longer-term studies were designed to expand upon findings from the 6- and 12-month RCTs. In a 2-year RCT (n = 841) of patients with early-stage COPD (ie, GOLD stage 1 [mild] or 2 [moderate]) in China, tiotropium HandiHaler® 18 μg QD compared with placebo significantly improved lung function throughout the 2-year period (range of mean differences in FEV₁: before bronchodilator use = 0.127–0.169 L; after bronchodilator use = 0.071–0.133 L; p < 0.001 for both) [25]. In addition, in the Understanding Potential Long-Term Impacts on Function with Tiotropium (UPLIFT) trial (n = 5993), tiotropium compared with placebo significantly improved lung function (range of mean differences in FEV₁: before bronchodilator use = 0.087–0.103 L; after bronchodilator use = 0.047–0.065 L; p < 0.001 for both), reduced the number of exacerbations/PY (0.73 vs 0.85; relative risk = 0.86 [ie, 14% reduction with tiotropium]; p < 0.001), increased the time to first exacerbation (p < 0.001), improved HRQoL (mean difference in SGRQ total score = − 2.7 U; p < 0.001), and reduced mortality (HR = 0.87; 95% confidence interval [CI] = 0.76–0.99) in patients with moderate-to-very severe COPD treated for 4 years [26]. Tiotropium significantly improved FEV₁ and HRQoL compared with placebo throughout the trial.

Conflicting results were reported, however, with respect to the effect of tiotropium on annual decline in FEV₁. Tiotropium compared with placebo significantly reduced the annual decline in FEV₁ after bronchodilator use in the aforementioned 2-year RCT in China [25] (mean decline in FEV₁ from day 30 to month 24: before bronchodilator use = 0.038 L/year vs 0.053 L/year [p = 0.06]; after bronchodilator use = 0.029 L/year vs 0.051 L/year [p = 0.006]) and in a retrospective analysis of two 1-year RCTs (mean decline in trough FEV₁ from days 8 to 344 = 0.012 vs 0.058 L/year; p = 0.005) [59]. In UPLIFT, however, the annual decline in FEV₁ was not significantly different between the tiotropium and placebo groups (mean decline in FEV₁ from day 30 to month 48: before bronchodilator use = 0.030 L/year vs 0.030 L/year [p = 0.95]; after bronchodilator use = 0.040 L/year vs 0.042 L/year [p = 0.21]) [26]. In a pre-specified subgroup analysis of the UPLIFT trial in patients with GOLD stage II COPD, the annual decline in post-bronchodilator FEV₁ was lower in the tiotropium group compared with the placebo group (mean decline in post-bronchodilator FEV₁ from day 30 to month 48: 0.043 L/year vs 0.049 L/year [p = 0.024]), supporting benefits of early intervention in COPD [60].

Long-term effects of tiotropium as first-line maintenance medication in COPD were further evaluated in a secondary analysis of data from the UPLIFT trial [61]. In addition, various post hoc and subgroup analyses were conducted to assess the long-term efficacy of 4 years of treatment with tiotropium with regard to smoking status [62], sex [63], and baseline FEV₁ ≥ 60% predicted [64]. In the latter analysis (n = 1210), tiotropium compared with placebo significantly improved lung function (difference in pre-bronchodilator FEV₁ = 0.087–0.127 L [range over months 1–48]; difference in post-bronchodilator FEV₁ = 0.038–0.084 L [range over months 1–48]; p ≤ 0.002 for both) and HRQoL (difference in SGRQ total score = − 2.0 to − 3.4 U; p < 0.05) and reduced the risk of exacerbations (HR = 0.83; 95% CI = 0.71–0.96) and mortality (HR = 0.66; 95% CI = 0.45–0.96) over the 4-year treatment period [64]. In another analysis of the UPLIFT data, patients taking ICS had significantly higher incidence rates of pneumonia (0.068 vs 0.056; p = 0.012) and COPD exacerbations (0.88 vs 0.62; p < 0.001) than patients not taking ICS [65]. An attenuated rate of pneumonia was observed in the tiotropium subgroup, in general, irrespective of ICS use (pneumonia incidence rate: fluticasone/placebo = 0.081; fluticasone/tiotropium = 0.073; other ICS/placebo = 0.062; other ICS/tiotropium = 0.055; no ICS/placebo = 0.055; and no ICS/tiotropium = 0.056), suggesting the adverse effects associated with the use of ICS might be mitigated with add-on tiotropium.

The only available SMI, Respimat®, is a pocket-sized, propellant-free device that generates an aerosol from a drug solution effectively and consistently [76]. Respimat®, which was developed to overcome or minimize some of the limitations of other inhalers [77], uses mechanical energy to generate a fine, slow-moving aerosol cloud (mist) from the drug solution, requires minimal coordination of inhalation and actuation, and does not require high inspiratory flow rates [14]. Further, the fine-particle fraction (approximately 75%; particles > 1 to < 5.8 μm) of the mist is nearly twice that in the aerosol cloud emitted from most pMDIs and DPIs [76], and the mist has a lower velocity and a longer duration than the aerosol cloud of pMDIs (mean velocity at a 10-cm distance from the nozzle: SMI = 0.8 m/s; pMDIs = 2.0–8.4 m/s; mean duration: SMI = 1.5 s; pMDIs = 0.15–0.36 s) [20]. These factors contribute to reduced oropharyngeal drug deposition and increased drug deposition in the lungs [15‐17].

Because Respimat® provides high drug lung deposition, lower doses of tiotropium were expected to provide comparable efficacy as the doses used with HandiHaler® [30]. A 3-week, dose-ranging RCT (n = 202) was conducted to compare various tiotropium doses (1.25 μg, 2.5 μg, 5 μg, 10 μg, or 20 μg QD) delivered via Respimat®, placebo delivered via Respimat®, tiotropium 18 μg delivered via HandiHaler®, and placebo delivered via HandiHaler® [30]. Tiotropium Respimat® 5 μg and 20 μg significantly improved lung function compared with placebo Respimat® (difference in trough FEV₁ response = 0.17 L [both doses]; p < 0.05 for both) at the end of treatment. Tiotropium Respimat® 10 μg also improved lung function but not significantly (p = 0.06) more than placebo [30]. Safety was similar across treatment groups.

In a pre-specified pooled analysis of two 30-week studies (n = 207), tiotropium Respimat® 5 μg and 10 μg QD were superior to placebo in improving lung function (day 29; difference in trough FEV₁ response = 0.126 L and 0.127 L, respectively; p < 0.0001 for both) and were non-inferior to tiotropium HandiHaler® 18 μg QD (day 29; difference in trough FEV₁ response = 0.029 L and 0.031 L, respectively; p < 0.0001 for both) [78]. In two identical, short-term, randomized, active- and placebo-controlled trials (n = 719), the efficacy and safety of tiotropium Respimat® 5 μg and 10 μg QD were compared with that of ipratropium 36 μg QID delivered via a pMDI [27]. At week 12, both tiotropium doses significantly improved lung function compared with ipratropium (difference in trough FEV₁ response = 0.064 L [p = 0.006] and 0.095 L [p < 0.001], respectively; difference in trough FVC response = 0.077 L [p > 0.01] and 0.125 L [p < 0.01], respectively) and compared with placebo (difference in trough FEV₁ response = 0.118 L and 0.149 L, respectively [both p < 0.0001]; difference in trough FVC response = 0.132 L [p < 0.01] and 0.180 L [p < 0.0001], respectively) [27]. Both tiotropium doses also significantly reduced rescue medication use compared with placebo (p = 0.0061 and p < 0.0001, respectively); however, only the 10-μg dose was statistically superior to ipratropium (p = 0.04). The long-term efficacy and safety of tiotropium Respimat® 5 μg and 10 μg QD were evaluated in two identical RCTs (n = 1990) [28]. At 1 year of treatment, both tiotropium doses compared with placebo significantly improved lung function (difference in trough FEV₁ response = 0.127 L and 0.150 L, respectively; both p < 0.0001) and HRQoL (difference in SGRQ total score = − 3.5 and – 3.8, respectively; both p < 0.0001) and reduced dyspnea (difference in TDI focal score = 1.05 and 1.08, respectively; both p < 0.0001) and mean COPD exacerbation rate/PY (odds ratio [OR] = 0.75 [p < 0.01] and 0.74 [p < 0.001], respectively). Significantly fewer patients experienced ≥1 exacerbation in the tiotropium groups than in the placebo group (difference = − 6.9 and − 7.2%, respectively; both p < 0.01). The incidence of gastrointestinal disorders was greater in the tiotropium groups than in the placebo group (dry mouth: 5 μg = 7.2%, 10 μg = 14.5%, placebo = 2.1%; constipation: 5 μg = 2.1%, 10 μg = 2.2%, placebo = 1.5%). In another 1-year RCT (n = 3991), the efficacy and safety of tiotropium Respimat® 5 μg QD were evaluated [29]. Patients were permitted to use usual therapy (any concurrent COPD medications except inhaled anticholinergics) to reflect closely the place of tiotropium in COPD management. At week 48, tiotropium 5 μg compared with placebo significantly improved lung function (adjusted mean difference in trough FEV₁ = 0.102 L; adjusted mean difference in trough FVC = 0.168 L; both p < 0.0001) and improved HRQoL (adjusted mean difference in SGRQ total score = − 2.9 U; p < 0.0001). Further, the proportion of patients having ≥1 exacerbation was lower in the tiotropium group compared with the placebo group (35.3 and 43.1%, respectively; HR = 0.69 [ie, 31% reduction in the risk of exacerbations with tiotropium; p < 0.0001) [29]. Collectively, results of these studies showed that tiotropium Respimat® 5 μg QD was efficacious as maintenance treatment for COPD and for reducing exacerbations, leading to the approval of Spiriva® Respimat® in the EU (2007), US (2014), and other countries [79].

The effectiveness of tiotropium Respimat® 5 μg QD was evaluated in a 6-week open-label observational study (n = 1230) [80]. Tiotropium improved physical function as indicated by a significant increase in the mean Physical Function subdomain (PF-10) score (baseline, 49.0 points; week 6, 62.3 points; difference, 13.4 points [p < 0.001]). Further, the proportion of patients achieving a ≥10-point improvement in PF-10 score was not significantly different between smokers and non-smokers (61.4 and 61.6%, respectively; difference, p = 0.93). Adverse events were reported by 4.0% of patients, the most common being respiratory symptoms and dry mouth.

The TIOtropium Safety and Performance In Respimat® (TIOSPIR®) trial (n = 17,315) was conducted to confirm the safety and efficacy of tiotropium Respimat® in a large population of patients with COPD [81]. By having an adequate patient population size (> 17,000 patients at > 1200 investigator sites in 50 countries), broad inclusion criteria to closely reflect real-world patients with COPD, and sufficient treatment duration, analyses of all-cause mortality and time to first COPD exacerbation were possible. During a 2.3-year mean follow-up period, tiotropium Respimat® 5 μg and 2.5 μg QD were non-inferior to tiotropium HandiHaler® 18 μg QD in the risk of death (both p < 0.05) and not superior in the risk of exacerbation (p = 0.42 and p = 0.56, respectively) [81]. In patients with a history of cardiac arrhythmia, tiotropium Respimat® 2.5 μg and tiotropium HandiHaler®18 μg had a similar impact on survival as measured by all-cause mortality. In addition to these analyses, various post hoc analyses were conducted to assess the impact of geographical variations [82] on COPD outcomes, to evaluate spirometry outcomes [83], and to determine risk factors for exacerbations [84]. Tiotropium Respimat® 5 μg was associated with a similar improvement in trough FVC and a comparable rate of decline in FEV₁ as tiotropium HandiHaler® 18 μg in the spirometry outcomes analysis [83]. In a multivariate analysis, baseline pulmonary maintenance medication was predictive of, and ICS use was associated with, increased exacerbation risk [84].

TIOSPIR® was conducted because a numerical, but not statistically significant, imbalance in the number of deaths had been noted in meta-analyses of Respimat® trials, between tiotropium Respimat® and placebo Respimat® or tiotropium HandiHaler®, particularly for patients with known cardiac disorders [85‐87]. Valid concerns were quickly raised, however, about the meta-analyses’ methodologies and conclusions, particularly the analysis conducted by Singh et al. [85], which included five of the aforementioned RCTs [27‐29, 88‐90]. For example, in that analysis, conclusions were based on three 1-year studies that had mortality imbalances (and not based on two 12-week studies that showed no imbalance) [88]. Additionally, in one of the studies with mortality imbalances, the placebo arm had an unusually low (0.77%) mortality rate compared with that observed in other studies (1.5–2.5% or higher) [88]. Further, fatal cases were incorrectly assigned to treatment groups [88, 91], and a 6-month trial of > 850 patients [92], which had only two deaths in the tiotropium Respimat® arm and five deaths in the placebo Respimat® arm, was not included [88, 91]. Moreover, two doses of tiotropium—the marketed 5-μg dose and a non-approved, unmarketed 10-μg dose—were included in the primary analysis [91, 93]. These and other factors provide a plausible explanation for the imbalances observed in the aforementioned meta-analyses. In TIOSPIR®, a similar impact on survival was observed between tiotropium Respimat® 2.5 μg and tiotropium HandiHaler® 18 μg, which expands upon findings from UPLIFT and reinforces the safety and efficacy of tiotropium in patients with COPD, regardless of cardiac arrhythmia history [26, 29, 81].

Lastly, results of an analysis of the tiotropium COPD clinical program (Table 2) showed that the characteristics of patients included in the RCTs were representative of “real-life” patient populations, thus demonstrating external validation of the results [94]. Further, tiotropium Respimat® 2.5 μg or 5 μg had similar exacerbation efficacy and safety to that of tiotropium HandiHaler® 18 μg [81].