Effect of Telmisartan on Walking Performance in Patients With Lower Extremity Peripheral Artery Disease

Key Points

Question

Does telmisartan improve walking ability in patients with lower extremity peripheral artery disease compared with placebo?

Findings

This randomized clinical trial included 114 participants with lower extremity peripheral artery disease randomized in a 2 × 2 factorial design to telmisartan plus exercise, telmisartan plus attention control, placebo plus exercise, or placebo plus attention control for 6 months. The between-group difference in change in 6-minute walk distance at 6-month follow-up for all participants randomized to telmisartan vs all participants randomized to placebo was −16.8 m, favoring placebo. This difference did not meet statistical significance.

Meaning

These results do not support telmisartan for improving walking performance in patients with lower extremity peripheral artery disease.

Abstract

Importance

Patients with lower extremity peripheral artery disease (PAD) have reduced lower extremity perfusion, impaired lower extremity skeletal muscle function, and poor walking performance. Telmisartan (an angiotensin receptor blocker) has properties that reverse these abnormalities.

Objective

To determine whether telmisartan improves 6-minute walk distance, compared with placebo, in patients with lower extremity PAD at 6-month follow-up.

Design, Setting, and Participants

Double-blind, randomized clinical trial conducted at 2 US sites and involving 114 participants. Enrollment occurred between December 28, 2015, and November 9, 2021. Final follow-up occurred on May 6, 2022.

Interventions

The trial randomized patients using a 2 × 2 factorial design to compare the effects of telmisartan plus supervised exercise vs telmisartan alone and supervised exercise alone and to compare telmisartan alone vs placebo. Participants with PAD were randomized to 1 of 4 groups: telmisartan plus exercise (n = 30), telmisartan plus attention control (n = 29), placebo plus exercise (n = 28), or placebo plus attention control (n = 27) for 6 months. The originally planned sample size was 240 participants. Due to slower than anticipated enrollment, the primary comparison was changed to the 2 combined telmisartan groups vs the 2 combined placebo groups and the target sample size was changed to 112 participants.

Main Outcomes and Measures

The primary outcome was the 6-month change in 6-minute walk distance (minimum clinically important difference, 8-20 m). The secondary outcomes were maximal treadmill walking distance; Walking Impairment Questionnaire scores for distance, speed, and stair climbing; and the 36-Item Short-Form Health Survey physical functioning score. The results were adjusted for study site, baseline 6-minute walk distance, randomization to exercise vs attention control, sex, and history of heart failure at baseline.

Results

Of the 114 randomized patients (mean age, 67.3 [SD, 9.9] years; 46 were women [40.4%]; and 81 were Black individuals [71.1%]), 105 (92%) completed 6-month follow-up. At 6-month follow-up, telmisartan did not significantly improve 6-minute walk distance (from a mean of 341.6 m to 343.0 m; within-group change: 1.32 m) compared with placebo (from a mean of 352.3 m to 364.8 m; within-group change: 12.5 m) and the adjusted between-group difference was −16.8 m (95% CI, −35.9 m to 2.2 m; P = .08). Compared with placebo, telmisartan did not significantly improve any of the 5 secondary outcomes. The most common serious adverse event was hospitalization for PAD (ie, lower extremity revascularization, amputation, or gangrene). Three participants (5.1%) in the telmisartan group and 2 participants (3.6%) in the placebo group were hospitalized for PAD.

Conclusions and Relevance

Among patients with PAD, telmisartan did not improve 6-minute walk distance at 6-month follow-up compared with placebo. These results do not support telmisartan for improving walking performance in patients with PAD.

Trial Registration

ClinicalTrials.gov Identifier: NCT02593110

This randomized clinical trial compares the effect of telmisartan on 6-minute walk distance compared with placebo in patients with lower extremity peripheral artery disease at 6-month follow-up.

Introduction

Few noninvasive therapies improve walking impairment in patients with lower extremity peripheral artery disease (PAD).¹ In addition to reduced lower extremity perfusion, patients with PAD have damaged lower extremity skeletal muscle characterized by reduced muscle area, increased muscle fibrosis, and impaired mitochondrial structure and function.^2,3 Therapies that improve lower extremity perfusion and repair ischemia-related skeletal muscle damage may have potential for improving walking performance in patients with PAD.⁴

During a preclinical study,⁵ the angiotensin II receptor blockade promoted muscle regeneration and reduced fibrosis in damaged skeletal muscle. Angiotensin receptors type 1 and 2 modulate the skeletal muscle microcirculatory system, promoting vasoconstriction and vasodilation, respectively.⁶ During a preclinical study,⁷ the angiotensin receptor blocker (ARB) losartan preferentially blocked angiotensin receptor 1, thereby increasing skeletal muscle perfusion. Telmisartan has an affinity for angiotensin receptor 1 that is more than 3000-fold greater than its affinity for angiotensin receptor 2, indicating that telmisartan may substantially increase perfusion in the microcirculatory system.⁸ In a randomized clinical trial of 36 patients with PAD,⁹ telmisartan significantly increased treadmill walking distance and brachial artery flow-mediated dilation at 12-month follow-up compared with placebo.

The Telmisartan Plus Exercise to Improve Functioning in Peripheral Artery Disease (TELEX) trial randomized patients using a 2 × 2 factorial design to test whether telmisartan improved 6-minute walk distance more than placebo and whether telmisartan combined with supervised exercise improved 6-minute walk distance more than supervised exercise alone and more than telmisartan plus attention control in patients with PAD.

Methods

The TELEX randomized clinical trial involved 2 sites in the US (Northwestern University and Tulane University). The institutional review boards at each site approved the trial protocol and statistical analysis plan (Supplement 1). Participants gave written informed consent. The trial randomized patients using a 2 × 2 factorial design to 4 parallel groups for 6 months: telmisartan plus exercise, telmisartan plus attention control, placebo plus exercise, or placebo plus attention control. The trial was double-blinded regarding telmisartan and placebo. Outcome assessors were unaware of participant assignment to exercise or attention control. Enrollment occurred between December 28, 2015, and November 9, 2021. Final follow-up occurred on May 6, 2022.

Most potential participants were already taking an ARB or angiotensin-converting enzyme (ACE) inhibitor, resulting in slower recruitment than anticipated. Therefore, in the fall of 2019, investigators requested a meeting with the data and safety monitoring board and sponsor and proposed modification of the primary aims to a single comparison of telmisartan (exercise and attention control groups) vs placebo (exercise and attention control groups), reducing the sample size needed for the primary outcome to 112 participants. The data and safety monitoring board and study sponsor approved this modification on October 31, 2019, with 102 participants enrolled of the originally planned sample size of 240. No efficacy data had been analyzed by the data and safety monitoring board, sponsor, or investigators when this decision was made. Recruitment continued until funding was no longer available.

Participant Identification

Participants were recruited using lists of patients with PAD from electronic health records and physician referrals at each site. Postcards advertising the study were mailed to patients aged 50 years or older in Chicago, Illinois, and New Orleans, Louisiana. Study advertisements were placed on public transportation in each city. Individuals with PAD who completed prior studies with the site principal investigators (M.M.M. and L.B.) and expressed interest in future research also were invited to participate.

Inclusion and Exclusion Criteria

The inclusion criterion was an ankle-brachial index (ABI) of 0.90 or less in either leg.¹⁰ Individuals with a resting ABI of greater than 0.90 and of 1.00 or less at baseline were eligible if their ABI decreased by 20% or greater after a heel-rise test.¹¹ Individuals with an ABI greater than 0.90 were eligible if (1) a hospital-based vascular laboratory report showed an ABI of less than 0.90 or a toe-brachial index of less than 0.70 in patients with ischemic leg symptoms or only a toe-brachial index of less than 0.60 in patients without ischemic leg symptoms or (2) an angiogram showed at least 70% stenosis in an artery supplying the legs.

Individuals taking an ACE inhibitor or ARB were excluded. Additional exclusion criteria included (1) having a systolic blood pressure of less than 100 mm Hg and a diastolic blood pressure of less than 50 mm Hg, (2) having a potassium level greater than 5.0 mmol/L, (3) having a major amputation, (4) being confined to a wheelchair, (5) using a walking aid other than a cane, (6) having a walking limitation for a reason other than PAD, (7) having a foot ulcer or critical limb ischemia, (8) having a Mini-Mental State Examination score of less than 23, (9) having severe visual or hearing impairment that interfered with trial participation, (10) undergoing lower extremity or coronary revascularization or major surgery during the previous 3 months, (11) having a planned revascularization or major surgery in the next 6 months, and (12) having a major medical illness such as cancer. In addition, individuals who were already exercising at a level similar to that targeted in the exercise intervention were excluded as well as those for whom exercise might be unsafe.

Potential participants were asked to take 20 mg of telmisartan daily for 14 days prior to randomization (run-in period).¹² Those who did not take telmisartan for at least 10 days were excluded. Those with new lightheadedness, a potassium level of 5.5 mmol/L or greater, a decrease in estimated glomerular filtration rate by 30% or greater, or a systolic blood pressure of less than 100 mm Hg and a diastolic blood pressure of less than 50 mm Hg at the end of the run-in period were excluded.

Measurement of ABI

A handheld Doppler probe (Pocket Dop II; Nicolet Biomedical, Inc) was used to measure systolic blood pressure twice in the right and left brachial, dorsalis pedis, and posterior tibial arteries.^10,13 The ABI was calculated by dividing the mean of the dorsalis pedis and posterior tibial artery pressures in each leg by the mean of the 4 brachial pressures.¹³

Medical History

Medical history, race, and demographics were obtained by questionnaire. To comply with funding agency requirements and to assess the generalizability of the results, participant self-reported information on race was collected using an open-ended question and classified with fixed categories.

Randomization

Eligible participants were randomized using a 2 × 2 factorial design to 1 of the 4 study groups (Figure 1) with the SAS computer program version 9.4 (SAS Institute Inc), and were stratified by study site according to whether the participant consented to a gastrocnemius muscle biopsy. Block randomization was used with block sizes randomly selected from 8 and 12.

Figure 1. Flow of Participants Included or Excluded in the Telmisartan Plus Exercise to Improve Functioning in Peripheral Artery Disease Trial.

ACE indicates angiotensin-converting enzyme; ARB, angiotensin receptor blocker.

^aThe inclusion criterion was a value of 0.90 or less in either leg. Individuals with resting values of greater than 0.90 and 1.00 or less at baseline were eligible if their values decreased by 20% or greater after a heel-rise test. Individuals with values of greater than 0.90 were eligible if (1) a hospital-based vascular laboratory report showed a value of less than 0.90 or a toe-brachial index of less than 0.70 in patients with ischemic leg symptoms or only a toe-brachial index of less than 0.60 in patients without ischemic leg symptoms or (2) an angiogram showed at least 70% stenosis in an artery supplying the legs.

Telmisartan and Placebo Interventions

Telmisartan and placebo were dispensed in doses of 20 mg/d, 40 mg/d, or 80 mg/d. Participants began taking their assigned pill at either a dose of 20 mg/d or 40 mg/d, which was determined by the principal investigator (M.M.M.) based on their blood pressure and potassium levels after the run-in period. Participants returned for blood pressure and potassium measurements 2 weeks after each dose increase. Doses were titrated to a maximum of 80 mg/d as tolerated. Participants also returned monthly for blood pressure and potassium measures at the 2-month and 5-month follow-up visits.

Supervised Exercise and Attention Control

Walking exercise on a treadmill was administered 3 times weekly with an exercise physiologist, working up to 50 minutes of exercise per session.^14,15,16 Participants randomized to attention control attended weekly 1-hour educational sessions by faculty or staff on health topics including cancer screening, immunizations, and hypertension.^14,15,16

Primary Outcome

The primary outcome was the 6-month change in 6-minute walk distance (minimum clinically important difference [MCID], 8-20 m).¹⁷

Secondary Outcomes

The secondary outcomes were the 6-month changes in maximal treadmill walking distance, Walking Impairment Questionnaire (WIQ) scores for distance, speed, and stair climbing (score range, 0-100; 100 indicates the best score), and the 36-Item Short Form Health Survey (SF-36) physical functioning score (range, 0-100; 100 indicates the best score).^18,19,20

Exploratory Outcomes

The exploratory outcomes were 6-month changes in usual- and fast-paced 4-m walking velocity and Short Physical Performance Battery (SPPB) score (range, 0-12; 12 indicates the best score).^14,21 The exploratory outcomes not presented in this article were 6-month change in gastrocnemius muscle biopsy measures (satellite cell number; type I muscle fiber composition; peroxisome proliferator activated receptor [PPAR] δ; PPAR γ coactivator 1α; adenosine monophosphate–activated protein kinase quantity and activity; and the relative gene expression of PPAR δ, PPAR γ coactivator 1α, adenosine monophosphate–activated protein kinase, FOX03A, and SIRT1).

Descriptions of Outcome Measures

6-Minute Walk Distance

Participants walked up and down a 30-m hallway after a script was read instructing them to cover as much distance as possible within 6 minutes.^14,15,16,17 The distance completed after 6 minutes was recorded. The MCID ranges from 8 m to 20 m in patients with PAD.^22,23

Maximal Treadmill Walking Distance

Treadmill walking performance was measured using the Gardner-Skinner protocol.¹⁸ No MCID has been defined for change in treadmill walking distance in patients with PAD.

The WIQ

The WIQ is a PAD-specific measure of self-reported walking limitations (score range, 0-100; 100 indicates the best score).¹⁹ The WIQ distance score measures difficulty walking distances up to 450 m. The WIQ speed score measures difficulty walking varying speeds for 1 block, ranging from slowly to jogging. The WIQ stair climbing score measures difficulty climbing stairs. Small MCID values have been defined as 5 points, 2 points, and 12 points for the WIQ distance, speed, and stair climbing scores, respectively.²³

Health-Related Quality of Life

The SF-36 physical functioning score measured health-related quality of life (score range, 0-100; 100 indicates the best score).²⁰ The MCID values for the SF-36 physical functioning score range from 5 points to 7 points.²⁴

The SPPB

The SPPB combines data from usual-paced 4-m walking velocity, time to rise from a seated position 5 times, and standing balance.^14,21,25 Individuals receive a score of 0 for each task they are unable to complete. Scores from 1 to 4 are assigned for the remaining tasks according to established methods and are summed to obtain the SPPB score (range, 0-12; 12 indicates the best score).^14,21,25

4-m Walking Velocity

Walking velocity was measured over 4 m at the usual pace and at the fastest pace using previously described methods.^14,21,25

Power Calculation

Accounting for patients lost to follow-up, 25 participants were anticipated in each of the 4 groups at 6-month follow-up, providing 80% power to detect a minimum difference of 0.566 SDs (ie, 30 m) between the telmisartan and placebo groups for change in 6-minute walk distance using a 2-sided, 2-sample t test with a significance level of .05. This power calculation was performed when the primary aim was changed to a single 2-group comparison before any efficacy data were reviewed.

Statistical Analyses

The baseline characteristics were summarized using means (SDs) for continuous variables and counts (proportions) for categorical variables. All participants who completed follow-up were analyzed according to their originally randomized group, irrespective of whether they adhered to their assigned interventions. For the primary outcome, analysis of covariance was used because of its greater statistical power than the t test to compare changes in 6-minute walk distance at 6-month follow-up between the telmisartan and placebo groups, adjusting for baseline 6-minute walk distance, study site, randomization to exercise, sex, and history of heart failure at baseline. Sex and history of heart failure were selected before any outcome data were analyzed because the prevalence of these characteristics was not balanced at baseline across the 4 groups.

Similar analyses were performed to compare changes in the secondary and exploratory outcomes between participants randomized to telmisartan and those randomized to placebo and were adjusted for baseline 6-minute walk distance, sex, heart failure, and study site. The statistical significance of all tests was set at a 2-sided level of .05. Because of the potential for type I error due to multiple comparisons, the findings for the analyses of the secondary outcomes should be interpreted as exploratory. In the post hoc analyses, change in 6-minute walk distance was compared between those randomized to placebo plus attention control and those randomized to exercise alone. The statistical analyses were performed using SAS version 9.4 (SAS Institute Inc).

Results

Of 244 patients with PAD who gave written informed consent, 114 were randomized (mean age, 67.3 years [SD, 9.9 years]; 46 were women [40.4%]; and there were 81 Black individuals [71%]; Table 1) and 105 (92%) completed 6-month follow-up (Figure 1).

Table 1. Baseline Characteristics of Participants With Peripheral Artery Disease.

	Main comparison		Original comparison groups
	Telmisartan	Placebo	Telmisartan + exercise	Telmisartan + attention control	Placebo + exercise	Placebo +  attention control
No. of participants	59	55	30	29	28	27
Age, mean (SD), y	67.7 (9.4)	66.8 (10.4)	68.5 (8.9)	66.9 (10.0)	65.8 (9.4)	67.9 (11.5)
Sex, No. (%)
Male	37 (62.7)	31 (56.4)	18 (60.0)	19 (65.5)	13 (46.4)	18 (66.7)
Female	22 (37.2)	24 (43.6)	12 (40)	10 (34.4)	15 (53.5)	9 (33.3)
Race, No. (%)
American Indian/Alaska Native	1 (1.7)	0	0	1 (3.5)	0	0
Asian	1 (1.7)	0	1 (3.3)	0	0	0
Black	43 (72.9)	38 (69.1)	22 (73.3)	21 (72.4)	18 (64.3)	20 (74.1)
White	11 (18.6)	15 (27.3)	6 (20.0)	5 (17.2)	9 (32.1)	6 (22.2)
Other, unknown, or not reported	3 (5.1)	2 (3.6)	1 (3.3)	2 (6.9)	1 (3.6)	1 (3.7)
Ankle-brachial index, mean (SD)	0.72 (0.27)	0.74 (0.21)	0.67 (0.23)	0.77 (0.30)	0.78 (0.23)	0.69 (0.17)
Body mass index, mean (SD)a	29.1 (7.0)	28.5 (6.0)	30.5 (7.5)	27.6 (6.2)	28.7 (6.4)	28.2 (5.8)
History, No. (%)
Current smoker	30 (50.9)	28 (50.9)	14 (46.7)	16 (55.2)	14 (50.0)	14 (51.9)
Diabetes	15 (25.4)	19 (34.6)	10 (33.3)	5 (17.2)	9 (32.1)	10 (37.0)
Pulmonary disease	14 (23.7)	12 (21.8)	5 (16.7)	9 (31.0)	8 (28.6)	4 (14.8)
Cancer	8 (13.6)	12 (21.8)	3 (10.0)	5 (17.2)	7 (25.0)	5 (18.5)
Myocardial infarction	8 (13.6)	9 (16.4)	3 (10.0)	5 (17.2)	4 (14.3)	5 (18.5)
Stroke	8 (13.6)	6 (10.9)	6 (20.0)	2 (6.9)	4 (14.3)	2 (7.4)
Angina	3 (5.1)	5 (9.1)	3 (10.0)	0	3 (10.7)	2 (7.4)
Heart failure	1 (1.7)	6 (10.9)	1 (3.3)	0	1 (3.6)	5 (18.5)
Leg symptoms, No. (%)
Classic intermittent claudication	4 (6.8)	9 (16.4)	2 (6.7)	2 (6.9)	4 (14.3)	5 (18.5)
Exertional other than classic intermittent claudication	49 (83.1)	42 (76.4)	25 (83.3)	24 (82.8)	22 (78.6)	20 (74.1)
No exertional	6 (10.2)	4 (7.3)	3 (10.0)	3 (10.3)	2 (7.1)	2 (7.4)
Medication use, No. (%)
Calcium channel blocker	25 (42.4)	9 (16.4)	15 (50.0)	10 (34.5)	5 (17.9)	4 (14.8)
β-Blocker	20 (33.9)	17 (30.9)	11 (36.7)	9 (31.0)	5 (17.9)	12 (44.4)
Diuretic	11 (18.6)	10 (18.2)	6 (20.0)	5 (17.2)	7 (25.0)	3 (11.1)
Statin	35 (59.3)	34 (61.8)	17 (56.7)	18 (62.1)	17 (60.7)	17 (63.0)
Metformin	6 (10.2)	9 (16.4)	5 (16.7)	1 (3.5)	5 (17.9)	4 (14.8)
Insulin	4 (6.8)	4 (7.3)	3 (10.0)	1 (3.5)	2 (7.1)	2 (7.4)
6-min walk distance, m
Mean (SD)	343.2 (91.5)	345.5 (125.1)	327.3 (96.6)	359.6 (84.4)	317.4 (123.3)	374.6 (122.4)
Median (IQR)	355.1 (292.9-402.3)	381.9 (262.1-432.8)	349.6 (258.5-378.9)	355.4 (327.1-420.9)	339.4 (228.3-419.6)	401.4 (291.1-460.9)
Maximal treadmill walking distance, mb
Mean (SD)	389.6 (248.4)	417.3 (268.8)	320.0 (191.6)	461.7 (281.5)	388.6 (239.4)	447.0 (297.8)
Median (IQR)	359.4 (178.8-543.6)	430.1 (167.1-600.8)	295.5 (149.3-475.7)	447.9 (232.5-568.6)	398.8 (163.4-568.2)	445.3 (167.2-712.6)
4-m walking velocity, mean (SD), m/s
Usual paced	0.84 (0.17)	0.86 (0.21)	0.82 (0.19)	0.86 (0.14)	0.80 (0.21)	0.92 (0.20)
Fast paced	1.12 (0.22)	1.14 (0.29)	1.09 (0.23)	1.16 (0.20)	1.03 (0.26)	1.26 (0.28)
Short Physical Performance Battery, mean (SD)c	9.69 (2.06)	9.58 (2.26)	9.46 (2.01)	9.92 (2.13)	9.18 (2.60)	10.04 (1.74)

^a Calculated as weight in kilograms divided by height in meters squared.

^b The maximal time can range from less than 2 minutes to more than 25 minutes. For most participants, the speed of the treadmill was set at 2.0 mph and the treadmill grade was increased by 2% every 2 minutes. Individuals unable to walk at 2.0 mph were started at a treadmill speed of 0.50 mph and the speed was increased by 0.50 mph every 2 minutes until the speed reached 2.0 mph, at which point the grade was increased by 2% every 2 minutes.

^c The score range is 0 to 12; 12 indicates the best score.

Intervention Adherence and Blood Pressure Values

Adherence based on pill counts was 92.6% for telmisartan and 90.4% for placebo. Three participants (5%) randomized to telmisartan and 2 participants (3.6%) randomized to placebo stopped taking their study medication due to a decrease in blood pressure, a decrease in estimated glomerular filtration rate greater than 30%, or new lightheadedness. In addition, 4 participants (6.7%) in the telmisartan group and 5 participants (9%) in the placebo group stopped taking the study drugs for personal reasons (n = 7), because their physician prescribed them an ACE inhibitor (n = 1), and because their physician advised them to discontinue the study drug (n = 1). Systolic, diastolic, and mean arterial pressures were lower after baseline in participants randomized to telmisartan compared with placebo (Figure 2 and eFigure in Supplement 2). The mean dose of study drug was 49.8 mg/d in the telmisartan group and 47.1 mg/d in the placebo group. After excluding participants who discontinued taking the study drugs, the mean dose was 53.4 mg/d in the telmisartan group and 50.3 mg/d in the placebo group.

Figure 2. Changes in Mean Arterial Pressure.

The horizontal line within the box represents the median value and the dot within the box represents the mean value. The top end of the box represents the 75th quartile and the bottom of the box represents the 25th quartile. The very top of the vertical line, indicated by the short horizontal bar, represents the maximum value. The very bottom of the vertical line, indicated by the short horizontal bar, represents the minimum value. Dots outside the box plots represent outlier values.

Primary Outcome

At 6-month follow-up, telmisartan did not significantly improve 6-minute walk distance (from a mean of 341.6 m [SD, 92.8 m] at baseline to 343.0 m [SD, 92.7 m] at 6-month follow-up [within-group change, 1.32 m]) compared with placebo (from a mean of 352.3 m [SD, 116.1 m] to 364.8 m [SD, 107.0 m] [within-group change, 12.5 m]) and the adjusted mean between-group difference was −16.8 m (95% CI, −35.9 to 2.2 m; P = .08) (Table 2), trending toward an adverse effect of telmisartan on walking performance (Figure 3).

Table 2. Primary and Secondary Outcomes.

	Telmisartan (n = 59)			Placebo (n = 55)			Telmisartan vs placebo
	Baseline	6-mo follow-up	Within-group change, mean (SD)a	Baseline	6-mo follow-up	Within-group change, mean (SD)a	Adjusted between-group difference, mean (95% CI)b	P value
Primary outcome
6-min walk distance, mc
Mean (SD)	(n = 56) 341.6 (92.8)	(n = 56) 343.0 (92.7)	1.32 (58.5)	(n = 47) 352.3 (116.1)	(n = 47) 364.8 (107.0)	12.5 (40.4)	−16.8 (−35.9 to 2.2)	.08
Median (IQR)	354.8 (291.8 to 399.3)	342.0 (292.8 to 419.3)	10.4 (−27.0 to 28.7)	383.4 (274.3 to 432.8)	390.8 (303.9 to 432.8)	14.3 (−11.9 to 39.3)
Secondary outcomes
Maximal treadmill walking distance, md
Mean (SD)	(n = 50) 383.0 (229.3)	(n = 50) 523.0 (342.0)	139.9 (202.0)	(n = 39) 430.3 (265.6)	(n = 39) 557.9 (294.9)	127.6 (224.1)	5.2 (−85.1 to 95.5)	.91
Median (IQR)	367.5 (179.7 to 503.4)	448.8 (323.7 to 712.6)	107.3 (−2.7 to 239.6)	439.9 (167.2 to 600.8)	488.2 (330.8 to 766.2)	77.8 (3.6 to 305.8)
Walking Impairment Questionnaire distance scoree
Mean (SD)	(n = 55) 37.7 (30.6)	(n = 55) 45.3 (33.7)	7.6 (29.8)	(n = 48) 42.1 (32.1)	(n = 48) 45.8 (32.9)	3.6 (15.8)	4.2 (−5.7 to 14.1)	.40
Median (IQR)	26.6 (11.1 to 59.5)	37.1 (12.0 to 78.6)	3.3 (−5.7 to 19.0)	28.8 (14.0 to 70.0)	39.5 (22.0 to 70.2)	1.3 (−1.8 to 10.2)
Walking Impairment Questionnaire speed scoref
Mean (SD)	(n = 55) 39.4 (23.5)	(n = 55) 43.1 (27.4)	3.7 (25.8)	(n = 48) 40.5 (27.7)	(n = 48) 47.1 (25.8)	6.6 (17.5)	−2.1 (−11.2 to 7.1)	.66
Median (IQR)	39.1 (21.7 to 56.5)	43.5 (21.7 to 60.9)	1.1 (−13.0 to 15.2)	38.6 (18.5 to 57.6)	43.5 (27.2 to 63.0)	7.1 (−4.3 to 15.2)
Walking Impairment Questionnaire stair climbing scoreg
Mean (SD)	(n = 55) 53.9 (31.8)	(n = 55) 56.9 (30.0)	3.0 (31.7)	(n = 48) 53.2 (29.1)	(n = 48) 57.5 (28.8)	4.3 (23.4)	−1.3 (−12.7 to 10.1)	.83
Median (IQR)	54.2 (25.0 to 87.5)	54.2 (29.2 to 79.2)	0 (−12.5 to 16.7)	54.2 (25.0 to 77.1)	60.4 (41.7 to 77.1)	0 (0 to 14.6)
SF-36 physical functioning scoreh
Mean (SD)	(n = 55) 48.5 (21.5)	(n = 55) 51.7 (23.9)	3.2 (20.9)	(n = 48) 55.5 (24.5)	(n = 48) 60.0 (25.6)	4.5 (21.3)	1.2 (−7.3 to 9.6)	.79
Median (IQR)	45.0 (30.0 to 65.0)	55.0 (35.0 to 70.0)	5.0 (−5.0 to 10.0)	60.0 (35.0 to 75.0)	65.0 (40.0 to 80.0)	0 (−5.0 to 15.0)
Exploratory outcomes
4-m walking velocity, mean (SD), m/s
Usual paced	(n = 50) 0.85 (0.17)	(n = 50) 0.86 (0.16)	(n = 50) 0.01 (0.13)	(n = 39) 0.89 (0.20)	(n = 39) 0.93 (0.20)	(n = 39) 0.04 (0.14)	−0.04 (−0.10 to 0.02)	.17
Fast paced	(n = 50) 1.14 (0.21)	(n = 50) 1.16 (0.23)	(n = 50) 0.02 (0.16)	(n = 39) 1.18 (0.27)	(n = 39) 1.21 (0.24)	(n = 39) 0.03 (0.16)	−0.01 (−0.07 to 0.06)	.86
Short Physical Performance Battery, mean (SD)i	(n = 49) 9.76 (2.04)	(n = 49) 9.86 (1.86)	(n = 49) 0.10 (1.46)	(n = 38) 9.92 (2.08)	(n = 38) 10.21 (1.89)	(n = 38) 0.29 (1.86)	−0.17 (−0.89 to 0.55)	.63

Abbreviation: SF-36, 36-Item Short Form Health Survey.

^a Some of the data are median (IQR) as indicated in the rows.

^b Adjusted for study site, baseline 6-minute walk distance, randomization to exercise, sex, and history of heart failure at baseline.

^c Adjusted using analysis of covariance for baseline distance, sex, heart failure, study site, and randomization to exercise or attention control. Represents the maximal distance a participant can walk in 6 minutes and potentially ranges from a small number of meters to farther than 500 m. The minimum clinically important difference values have been defined as 8 m (small) to approximately 20 m (large).^18,19

^d The maximal time can range from less than 2 minutes to more than 25 minutes. For most participants, the speed of the treadmill was set at 2.0 mph and the treadmill grade was increased by 2% every 2 minutes. Individuals unable to walk at 2.0 mph were started at a treadmill speed of 0.50 mph and the speed was increased by 0.50 mph every 2 minutes until the speed reached 2.0 mph, at which point the grade was increased by 2% every 2 minutes.

^e Measures the participant’s reported difficulty in walking distances that range from across a small room to 450 m. Scores range from 0 to 100 (100 is the best score); higher scores indicate greater ease in walking long distances. Data were collected using a self-administered questionnaire. A minimum clinically important difference has not been defined.

^f Measures the participant’s reported difficulty in walking at different speeds ranging from slow to fast. Scores range from 0 to 100 (100 is the best score); higher scores indicate greater ease in walking faster speeds. Data were collected using a self-administered questionnaire. A minimum clinically important difference has not been defined.

^g Measures the participant’s reported difficulty in climbing stairs. Scores range from 0 to 100 (100 is the best score). Data were collected using a self-administered questionnaire. A minimum clinically important difference has not been defined.

^h Measures the participant’s perceived state of health and ability to function physically. Data were collected using a self-administered questionnaire. Scores range from 0 to 100; higher scores are better.

ⁱ The score range is 0 to 12; 12 indicates the best score.

Figure 3. Effects of Telmisartan on Change in 6-Minute Walk Distance in Patients With Peripheral Artery Disease.

A, Each vertical line represents an individual participant. Patients’ baseline values are ordered and graphed on the curved line. Each line extends to the patient’s 6-month value.

B, Each vertical line represents an individual participant. The vertical axis represents the 6-month change in 6-minute walk distance.

Secondary Outcomes

Compared with placebo, telmisartan did not significantly improve any of the 5 secondary outcomes at 6-month follow-up. The maximal treadmill walking distance changed from a mean of 383.0 m (SD, 229.3 m) at baseline to 523.0 m (SD, 342.0 m) at 6-month follow-up in the telmisartan group (within-group change, 139.9 m) compared with from a mean of 430.3 m (SD, 265.6 m) to 557.9 m (SD, 294.9 m) in the placebo group (within-group change, 127.6 m) and the adjusted mean between-group difference was 5.2 m (95% CI, −85.1 to 95.5 m; P = .91) (Table 2).

The WIQ scores and the SF-36 physical functioning scores were adjusted for baseline 6-minute walk distance, study site, sex, history of heart failure, and randomization to exercise or control group. The WIQ score for distance changed from a mean of 37.7 (SD, 30.6) at baseline to 45.3 (SD, 33.7) at 6-month follow-up in the telmisartan group (within-group change, 7.6) compared with from a mean of 42.1 (SD, 32.1) to 45.8 (SD, 32.9) in the placebo group (within-group change, 3.6) and the adjusted mean between-group difference was 4.2 (95% CI, −5.7 to 14.1; P = .40). The WIQ score for speed changed from a mean of 39.4 (SD, 23.5) at baseline to 43.1 (SD, 27.4) at 6-month follow-up in the telmisartan group (within-group change, 3.7) compared with from a mean of 40.5 (SD, 27.7) to 47.1 (SD, 25.8) in the placebo group (within-group change, 6.6) and the adjusted mean between-group difference was −2.1 (95% CI, −11.2 to 7.1; P = .66). The WIQ score for stair climbing changed from a mean of 53.9 (SD, 31.8) at baseline to 56.9 at 6-month follow-up (SD, 30.0) in the telmisartan group (within-group change, 3.0) compared with from a mean of 53.2 (SD, 29.1) to 57.5 (SD, 28.8) in the placebo group (within-group change, 4.3) and the adjusted mean between-group difference was −1.3 (95% CI, −12.7 to 10.1; P = .83). The SF-36 physical functioning score changed from a mean of 48.5 (SD, 21.5) at baseline to 51.7 (SD, 23.9) at 6-month follow-up in the telmisartan group (within-group change, 3.2) compared with from a mean of 55.5 (SD, 24.5) to 60.0 (SD, 25.6) in the placebo group (within-group change, 4.5) and the adjusted mean between-group difference was 1.2 (95% CI, −7.3 to 9.6; P = .79).

Exploratory Outcomes

There was no significant effect on change in any of the exploratory outcomes at 6-month follow-up. Usual-paced 4-m walking velocity changed from a mean of 0.85 m/s (SD, 0.17 m/s) at baseline to 0.86 m/s (SD, 0.16 m/s) at 6-month follow-up in the telmisartan group (within-group change, 0.01 m/s) compared with from a mean of 0.89 m/s (SD, 0.20 m/s) to 0.93 m/s (SD, 0.20 m/s) in the placebo group (within-group change, 0.04 m/s) and the adjusted mean between-group difference was −0.04 m/s (95% CI, −0.10 to 0.02 m/s; P = .17) (Table 2). Fast-paced 4-m walking velocity changed from a mean of 1.14 m/s (SD, 0.21 m/s) at baseline to 1.16 m/s (SD, 0.23 m/s) at 6-month follow-up in the telmisartan group (within-group change, 0.02 m/s) compared with from a mean of 1.18 m/s (SD, 0.27 m/s) to 1.21 m/s (SD, 0.24 m/s) in the placebo group (within-group change, 0.03 m/s) and the adjusted mean between-group difference was −0.01 m/s (95% CI, −0.07 to 0.06 m/s; P = .86). The SPPB score changed from a mean of 9.76 (SD, 2.04) at baseline to 9.86 (SD, 1.86) at 6-month follow-up in the telmisartan group (within-group change, 0.10) compared with from a mean of 9.92 (SD, 2.08) to 10.21 (SD, 1.89) in the placebo group (within-group change, 0.29) and the adjusted mean between-group difference was −0.17 (95% CI, −0.89 to 0.55; P = .63).

Post Hoc Analyses

In the post hoc analyses, compared with the placebo plus attention control group, exercise alone significantly improved 6-minute walk distance from a mean of 326.8 m at baseline to 357.0 m at 6-month follow-up in the exercise groups (within-group change, 30.3 m) compared with from a mean of 376.8 m to 372.2 m in the attention control groups (within-group change, −4.6 m) and the adjusted mean between-group difference was 21.3 m (95% CI, 1.4 to 41.1 m; P = .04).

Adverse Events

There were 16 serious adverse events in the telmisartan group and 21 in the placebo group. The most common serious adverse event was hospitalization for PAD (ie, lower extremity revascularization, amputation, or gangrene). Three participants (5.1%) in the telmisartan group and 2 participants (3.6%) in the placebo group were hospitalized for PAD.

Two serious adverse events were considered related to data collection. One participant developed a hematoma after the gastrocnemius muscle biopsy. One participant was hospitalized for severe hypertension identified during a study stress test.

Discussion

In this randomized clinical trial of patients with PAD, telmisartan had no significant effect on change in 6-minute walk distance; maximal treadmill walking distance; the WIQ scores for distance, speed, or stair climbing; and the SF-36 physical functioning score at 6-month follow-up compared with placebo. The effects of telmisartan trended toward a detrimental effect on 6-minute walk distance compared with placebo.

Clinical practice guidelines recommend ACE inhibitors or ARBs to reduce cardiovascular events in patients with PAD.²⁶ Knowledge of the effects of these drugs on walking performance will help physicians counsel patients with PAD about the expected effects of these drugs on their walking ability. If ARBs had a significantly detrimental effect on walking performance, some patients disabled by PAD may choose not to take an ARB for cardiovascular prevention, depending on their values and preferences.

In mice, compared with placebo, telmisartan increased type I muscle fibers, oxygen consumption, and running endurance.^27,28 In prior small clinical trials of patients with PAD, both telmisartan and ramipril, an ACE inhibitor, increased treadmill walking distance.^9,29 The current trial selected the 6-minute walk distance as the primary outcome because 6-minute walk distance better reflects walking performance in daily life than treadmill walking performance.¹⁷ In addition, a treadmill exercise intervention, such as that used in this trial, preferentially facilitates a learning effect for the treadmill walking outcome because patients with PAD become more comfortable walking on a treadmill during the treadmill walking sessions (3 times weekly), and a learning effect accounts for some of the improvement in treadmill walking distance.³⁰

There are several potential explanations for the lack of efficacy of telmisartan in this clinical trial. First, lower extremity atherosclerotic obstruction may have prevented sufficient delivery of telmisartan to skeletal muscle and microvasculature. Second, the effects of telmisartan on skeletal muscle may differ between animals and humans with PAD.^28,29 Third, the inclusion of supervised exercise in the 2 × 2 factorial design may have obscured a potential benefit from telmisartan. Fourth, it is possible that the modest blood pressure–lowering effect of telmisartan reduced perfusion or had other detrimental effects on walking performance. If correct, that could explain the point estimate (which was not statistically significant), suggesting a possible detrimental effect of telmisartan on 6-minute walk distance. Future studies should evaluate whether blood pressure–lowering medications may have detrimental effects in patients with PAD without high blood pressure.

The large proportion of potential participants with PAD already taking an ARB or ACE inhibitor prevented attainment of the originally planned sample size because use of an ARB or ACE inhibitor was an exclusion criterion. The widespread use of ARBs or ACE inhibitors in patients with PAD may reflect clinical practice guideline recommendations for ACE inhibitors or ARBs as a first-line therapy for treating hypertension in patients with PAD.^1,26 It is possible that patients with greater sensitivity to any benefits of ARBs were already taking an ARB or ACE inhibitor and therefore not eligible for this trial.

This study has several strengths. First, the trial was multicentered, increasing generalizability. Second, 71.1% of participants were Black individuals and PAD is common among Black patients.¹ Third, adherence to telmisartan was high. Fourth, the trial evaluated the effects of telmisartan on walking performance in patients engaged in exercise, which is a first-line therapy for PAD.^1,26

Limitations

This study had several limitations. First, the target sample size was reduced due to the large number of potential participants who were already taking an ACE inhibitor or ARB. Second, the clinical trial may have lacked statistical power to demonstrate a significant detrimental effect of telmisartan on walking performance in patients with PAD. Third, the results may not be generalizable to the patients who were excluded because they were already taking an ACE inhibitor or ARB. Fourth, baseline differences in history of heart failure, sex, and 6-minute minute walk distance required adjustment in the statistical analyses.

Conclusions

Among patients with PAD, telmisartan did not improve 6-minute walk distance at 6-month follow-up compared with placebo. These results do not support telmisartan for improving walking performance in patients with PAD.

Supplement 1.

Trial protocol

Supplement 2.

eFigure. Mean systolic and diastolic blood pressure values at monthly visits during the clinical trial among participants randomized to telmisartan or placebo

Supplement 3.

Data sharing statement