Changes in brachial artery endothelial function and resting diameter with moderate-intensity continuous but not sprint interval training in sedentary men

We compared the effects of 12 wk of moderate-intensity continuous training (MICT) and sprint interval training (SIT) on peripheral artery endothelial function and diameter, and central and lower limb stiffness in sedentary, healthy men. Whereas neither training program affected the popliteal artery or stiffness indexes, we observed changes in brachial artery function and diameter with MICT but not SIT. Brachial artery responses to SIT may follow a different time course or may not occur at all.

Abstract

Moderate-intensity continuous training (MICT) improves peripheral artery function in healthy adults, a phenomenon that reverses as continued training induces structural remodeling. Sprint interval training (SIT) elicits physiological adaptations similar to MICT, despite a lower exercise volume and time commitment; however, its effect on peripheral artery function and structure is largely unexplored. We compared peripheral artery responses to 12 wk of MICT and SIT in sedentary, healthy men (age = 27 ± 8 yr). Participants performed MICT (45 min of cycling at 70% peak heart rate; n = 10) or SIT (3 × 20-s “all out” cycling sprints with 2 min of recovery; n = 9), and responses were compared with a nontraining control group (CTL, n = 6). Allometrically scaled brachial flow-mediated dilation (FMD) increased 2.2% after 6 wk of MICT and returned to baseline levels by 12 wk, but did not change in SIT or CTL (group × time interaction, P = 0.04). Brachial artery diameter increased after 6 and 12 wk (main effect, P = 0.03), with the largest increases observed in MICT. Neither training protocol affected popliteal relative FMD and diameter, or central and lower limb arterial stiffness (carotid distensibility, central and leg pulse wave velocity) (P > 0.05 for all). Whereas earlier and more frequent measurements are needed to establish the potential presence and time course of arterial responses to low-volume SIT, our findings suggest that MICT was superior to the intense, but brief and intermittent SIT stimulus at inducing brachial artery responses in healthy men.

NEW & NOTEWORTHY We compared the effects of 12 wk of moderate-intensity continuous training (MICT) and sprint interval training (SIT) on peripheral artery endothelial function and diameter, and central and lower limb stiffness in sedentary, healthy men. Whereas neither training program affected the popliteal artery or stiffness indexes, we observed changes in brachial artery function and diameter with MICT but not SIT. Brachial artery responses to SIT may follow a different time course or may not occur at all.

the cardioprotective benefits of regular exercise are well established and include reduced risk for cardiovascular disease (CVD) (23) and improved arterial function and structure (15). Moderate-intensity continuous training (MICT), as reflected in physical activity guidelines (37), increases endothelial function (5, 36), reduces central arterial stiffness (17, 20), and elicits structural remodeling (36). Sprint interval training (SIT) and its companion, high-intensity interval training, have emerged as enjoyable and time-efficient alternatives to traditional MICT (11, 18) that could resolve the “lack of time” barrier for engaging in regular exercise (29). Interval training has been shown to elicit comparable, and at times superior, arterial responses than MICT (27); however, these observations have been made in clinical populations with compromised arterial health. In healthy adults, very few studies have examined the arterial responses to interval training, or compared its effect on arterial function and structure to MICT. To our knowledge, the effects of SIT on endothelial function have been examined in sedentary, healthy adults once (26) and only in the active lower limb. Furthermore, its effects on central stiffness are inconclusive, with studies reporting improvements (19) and no change (26).

Conventionally, brachial endothelial function is a surrogate for coronary endothelial function (1) and an independent predictor of CVD risk (14). Despite its predictive value, endothelial function in the brachial artery cannot be extrapolated to or from active lower limb arteries, as limb-specific differences have been observed (35). Previous studies have either examined the effects of MICT in both the nonactive upper limb and active lower limb (33, 36), or compared the effects of MICT and SIT, but only in the active lower limb (26). Our laboratory previously demonstrated that, in healthy young men and women, 6 wk of MICT or traditional Wingate-based SIT similarly improved endothelial function in the popliteal artery, but we did not assess the nonactive brachial artery (26). Furthermore, in our laboratory’s previous work, neither MICT nor SIT elicited changes in popliteal diameter; however, time course studies suggest that training first induces functional changes, followed by structural remodeling that returns function to baseline levels (21, 36).

No studies to date have directly compared the effects of MICT and SIT in both the nonactive and active peripheral arteries of sedentary, healthy men. Therefore, the aim of this study was to comprehensively examine the effects of 6 and 12 wk of MICT and low-volume SIT on brachial and popliteal artery endothelial function and diameter, and central and lower limb arterial stiffness in sedentary, healthy men compared with nontraining controls. We hypothesized that MICT and SIT would augment endothelial function in both the nonactive brachial artery and active popliteal artery after 6 wk of training, but that endothelial function would return to baseline levels by 12 wk. We further hypothesized that MICT and SIT would elicit comparable increases in brachial and popliteal diameter, and comparable improvements in arterial stiffness, that would manifest by 12 wk of training.

METHODS

Participants

Twenty-seven sedentary but otherwise healthy men between the ages of 19 and 44 yr were recruited through poster advertisement in and around McMaster University. Twenty-five participants completed the study, as one subject in each of the training groups withdrew from the study for unrelated reasons. The training-induced aerobic capacity and metabolic adaptations in the same subjects have been reported elsewhere by our collaborators (12). Participants were nonsmokers, with the exception of two former smokers in the control (CTL) group. All participants were classified as sedentary based on an International Physical Activity Questionnaire score of less than 600 metabolic equivalents (MET)-min/wk. Exclusion criteria included any known cardiovascular, cerebrovascular, respiratory, metabolic, musculoskeletal, neurological, or renal diseases, as well as the use of steroid or hormone therapy, erectile dysfunction medication, or long-term drug prescriptions, including statins. This study was approved by the Hamilton Integrated Research Ethics Board and conformed to the Declaration of Helsinki. Written, informed consent was obtained from the subjects before participation.

Study Design

Following an initial screening visit to verify eligibility, participants’ body mass index (kg/m²) and aerobic fitness were determined using anthropometric measures of height (m) and weight (kg), and an incremental peak oxygen uptake (V̇o_2peak) test on a cycle ergometer, respectively. Participants were assigned to a MICT (n = 10), SIT (n = 9), or nontraining control group (CTL, n = 6) based on age, body mass index, and V̇o_2peak to ensure equal distribution of these variables between groups at baseline. All assessments of V̇o_2peak were conducted on an electronically braked cycle ergometer (Lode Excalibur Sport, version 2.0, Groningen, The Netherlands), as described previously (13). Participants in the training groups were asked to refrain from engaging in additional forms of exercise training, while the control group was asked to remain sedentary. Training groups undertook a minimum of 30 supervised exercise sessions over a 12-wk period, with some participants completing up to 35 sessions to ensure a continued training stimulus until they could be scheduled for posttraining testing. Relevant to this publication, V̇o_2peak and vascular assessments were conducted in all groups at baseline and after 6 and 12 wk. Due to the sedentary nature of our participants at baseline, the training program incorporated a lead-in phase, with one session in week 1, two sessions in week 2, and three sessions per week thereafter, with the exception of week 7. Two exercise sessions in week 7 were replaced with testing visits. Vascular testing was conducted 24–48 h after a training session for the 6-wk assessment, and on average 7 days after the last training session for the 12-wk assessment.

Training Protocols

All training was conducted in the Human Performance Research Laboratory at McMaster University. Heart rate (HR) was monitored continuously during exercise in both groups (Polar A3, Lake Success, NY). Each session began with a 2-min warm-up and ended with a 3-min cool-down. The MICT protocol was modeled after the Canadian Physical Activity Guideline recommendations for adults of 150 min of moderate-to-vigorous intensity aerobic activity per week (37). MICT consisted of 45 min of continuous cycling (Ergo Race, Kettler, Ense-Parsit, Germany) at ~70% maximum HR (~55% of V̇o_2peak), for a total time commitment of 50 min. Mean HR for each 45-min MICT exercise bout was recorded, and workload was increased over the 12 wk to continue to elicit the same relative mean HR (70% maximum HR), thereby maintaining the same relative workload stimulus throughout the training program. SIT sessions consisted of 3 × 20-s “all out” cycling sprints (RacerMate; Velotron, Seattle, WA) against 5.0% body wt and interspersed with 2 min of active recovery at 50 W, for a total time commitment of 10 min (13). Sprint and recovery loads were set using compatible computer software (Wingate Software version 1.0; Velotron).

Vascular Assessments

Vascular assessments were conducted in the Vascular Dynamics Laboratory at McMaster University. Participants were tested in the morning after a 10-h overnight fast (no food or drink with the exception of water) and after having refrained from moderate-to-vigorous intensity activity for 24 h. All vascular assessments were conducted with the participant laying supine in a quiet, temperature-controlled room following a 10-min rest. Resting brachial blood pressure and HR were measured in triplicate using an automated oscillometric device (Dinamap Pro 100; Critikon, Tampa, FL), and the last two measurements were averaged. Resting hemodynamics were also monitored continuously throughout the testing session using a hemodynamic monitor (Finometer MIDI, Finapres Medical Systems; Amsterdam, The Netherlands) and single-lead ECG (model ML 132; ADInstruments, Colorado Springs, CO). For each visit, flow-mediated dilation (FMD) tests and resting arterial diameters were used to assess brachial and popliteal artery function and structure, respectively, while carotid distensibility and pulse-wave velocity (PWV) were used to assess arterial stiffness.

Flow-mediated dilation.

FMD is an index of endothelial-dependent vasodilation, with larger dilatory responses reflecting increased endothelial function. FMD was measured simultaneously in the right brachial and popliteal arteries in accordance with current guidelines (31). Blood pressure cuffs positioned around the forearm and calf were rapidly inflated to suprasystolic pressures (200 mmHg) for 5 min to occlude blood flow (BF) to the distal vascular beds. Two identical Doppler ultrasounds (Vivid Q; GE Medical Systems, Horten, Norway) were used to measure arterial diameter before cuff inflation (30-s image at rest) and from 5 s before cuff deflation to 3 min afterwards in both arteries. Longitudinal images (duplex mode) were obtained proximal to the antecubital fossa for the brachial artery, and proximal to the popliteal fossa for the popliteal artery. An insonation angle ≤68° was used for all scans to maximize image quality (25). End-diastolic frames were analyzed using semiautomated edge tracking software [Artery Measurement System (AMS) II, version 1.141; Gothenburg, Sweden] (38) to determine arterial diameter. Resting arterial diameter was used as a surrogate for arterial structure, and absolute and relative changes in FMD were calculated using Eqs. 1 and 2:

$$\text{FMD}\left(\text{mm}\right)=\text{peak RH diameter}-\text{resting diameter}$$ (1)

$$\text{FMD}\left(\text{\%}\right)=\frac{\text{absolute FMD}}{\text{resting diameter}}\times 100$$ (2)

Mean blood velocity (MBV) signals were also collected before cuff inflation (30 s at rest), and from 5 s before cuff deflation up to 3 min afterwards [reactive hyperemia (RH)]. For the brachial artery, an external spectral analysis system (model Neurovision 500M TCD; Multigon Industries, Yonkers, NY) was used to obtain continuous intensity weighted MBV, which was sampled using commercially available hardware (Powerlab model ML870, ADInstruments) and analyzed offline using compatible software (LabChart 7, ADInstruments). Popliteal MBV was analyzed using an available package on our ultrasound offline workstation (EchoPAC PC, version 110.0.2; GE Medical Systems). MBV (cm/s) and arterial diameter (cm) were averaged into five-cycle rolling bins for brachial and popliteal arteries and used to calculate BF (Eq. 3) and shear rate (SR, Eq. 4).

$$\text{BF}\left(\text{ml/min}\right)=\left(\text{π}{r}^2\times \text{MBV}\right)\times 60\hspace{0.17em},\text{\hspace{0.17em}where}r=(\text{diameter/2})$$ (3)

$$\text{SR}\left({\text{s}}^{-1}\right)=\frac{\text{MBV}\times 8}{\text{arterial diameter}}$$ (4)

We report the average resting BF and characterize the RH response by the peak postdeflation BF (peak RH BF), the MBV averaged to the peak RH diameter (MBV to peak), the SR area under the curve (AUC) to the peak RH diameter (SR AUC to peak), and the time to peak RH diameter (time to peak).

Common carotid artery distensibility.

Common carotid artery distensibility is the relative change in artery cross-sectional area for a given change in pressure, with larger values indicating reduced stiffness. Distensibility was measured using a combination of Doppler ultrasound (Vivid Q, GE Medical Systems) and applanation tonometry (model SPT-301, Millar Instruments, Houston, TX), as previously described (22). Briefly, a 12-MHz linear array ultrasound probe was used to obtain longitudinal images of the right common carotid artery (B-mode, 22.9 fps), while a tonometer was positioned over the strongest detectable pulse of the left common carotid artery to simultaneously obtain a waveform representing intra-arterial pressure. Signals were acquired for 10 consecutive cardiac cycles. For each cardiac cycle, ultrasound images were analyzed to determine maximum (d_max) and minimum lumen diameters (d_min), and tonometry signals were used to determine the pulse pressure (PP), defined as the difference between systolic and diastolic blood pressure. Distensibility was then calculated using Eq. 5:

$$\text{distensibility}\left({\text{mmHg}}^{-1}\right)=\frac{\text{π}{\left(\frac{d_{\text{max}}}{2}\right)}^2-\text{π}{\left(\frac{d_{\text{min}}}{2}\right)}^2}{\text{π}{\left(\frac{d_{\text{min}}}{2}\right)}^2\times \text{PP}}$$ (5)

Pulse wave velocity.

PWV is the speed of a pulse traveling along an arterial segment, with faster speeds reflecting stiffer arteries. PWV was assessed between the carotid and femoral arteries (central PWV) and between the femoral and dorsalis pedis arteries (leg PWV) and calculated using Eq. 6. The pulse travel distance was estimated between the two sites with an anthropometric measuring tape across the surface of the body, and 80% of the direct carotid-femoral distance was used to calculate central PWV, as recommended in the current guidelines (6). The pulse transit time was determined from pressure waveforms acquired with applanation tonometry (Mikro-Tip Catheter Transducer, model SPT-301; Millar Instruments), and digitally filtered (band pass, 5–30 Hz) to assist with the detection of the foot of each waveform, as previously described (22). PWV was analyzed in sets of 10 heart cycles and reported as the mean of 2 sets, or the median of 3 sets, if the difference between the first and second sets exceeded 0.5 m/s (6).

$$\text{PWV}\left(\text{m/s}\right)=\frac{\text{pulse travel distance}}{\text{pulse transit time}}$$ (6)

Statistical Analyses

All data were assessed for normality using the Shapiro-Wilk test. The impact of training on brachial and popliteal endothelial function and arterial diameter and arterial stiffness indexes was examined using a two-way mixed-model ANOVA, with factors of group (MICT, SIT, CTL) and time (baseline, 6 wk, 12 wk). Significant interactions or main effects were assessed with pairwise comparisons using Fisher’s least significant difference test. Where indicated, main effects were followed up with a one-way repeated-measures ANOVA. Statistical analyses were performed using SPSS Statistics (version 20.0, Chicago, IL) and GraphPad Prism (version 4.0b, La Jolla, CA). Significance was set at P ≤ 0.05, and data are expressed as means ± SD, unless otherwise noted.

Allometric Scaling of FMD

To determine whether changes in arterial diameter over the 12 wk necessitated allometric scaling of FMD, a linear regression analysis was used to determine the slope and 95% confidence intervals (95% CI) of the relationship between the natural log of peak RH diameter (lnD_peak, dependent variable) and resting diameter (lnD_rest, independent predictor) across all time points (baseline, 6 wk, 12 wk). Separate regression analyses were performed for each group (MICT, SIT, CTL). Allometric scaling of FMD has been recommended if the unstandardized β-coefficient deviates from 1 and/or the 95% CI has an upper limit <1, as this indicates peak RH diameter is not increasing as a constant proportion of resting diameter (3). When indicated by these criteria, we allometrically scaled FMD by using a linear mixed model with lnD_diff (i.e., lnD_peak − lnD_rest) as the dependent variable, group (MICT, SIT, CTL) and time (baseline, 6 wk, 12 wk) as fixed factors, and lnD_rest as the covariate. Fisher’s least significant difference test was used for pairwise post hoc comparisons (4). For each group at each time point, linear mixed-model estimated means (EM) were back transformed to obtain scaled FMD, [(e^EM − 1) × 100], and standard errors (SE) were back transformed and used to estimate standard deviations $\left\{\left[\left(e^{\text{SE}}–1\right)\hspace{0.17em}\times 100\right]\hspace{0.17em}\times \left(\sqrt{n}\right)\right\}$, where n is the group sample size. Scaled FMD for individual participants were estimated using the logged diameters and linear regression unstandardized β, $[\frac{\left(\ln D_{\mathit{\text{peak}}}\right)}{{\left(\ln D_{\mathrm{rest}}\right)}^{\beta }}-1]\times 100$.

RESULTS

Participants

Baseline characteristics for the 25 participants who completed the study are reported in Table 1. Due to difficulty imaging some participants during the FMD test, brachial data are reported for n = 24 (10 MICT, 8 SIT, 6 CTL), and popliteal data are reported for n = 22 (8 MICT, 8 SIT, 6 CTL). Due to unobtainable arterial pressure waveform signals in one participant, distensibility, central PWV, and leg PWV data are reported for n = 24 (9 MICT, 9 SIT, 6 CTL). There were no group differences at baseline for any of our measures (P > 0.05 for all).

Table 1.

Subject characteristics

	MICT	SIT	CTL
n	10	9	6
Age (range), yr	28 ± 9 (19–44)	27 ± 7 (19–39)	26 ± 8 (19–41)
Height, cm	176 ± 10	177 ± 11	176 ± 5
Weight, kg	84 ± 20	84 ± 23	78 ± 25
BMI, kg/m2	26 ± 6	27 ± 5	25 ± 7
V̇o2peak, ml·kg−1·min−1	34 ± 6	32 ± 7	32 ± 7

Values are means ± SD; n, no. of subjects. BMI, body mass index; V̇o_2peak, peak oxygen uptake. There were no group differences at baseline.

Training Effects

Participants in the MICT and SIT groups completed 32 ± 2 and 31 ± 1 supervised training sessions, respectively. V̇o_2peak increased similarly in both training groups by ~19% after 12 wk (P < 0.001 for both), but was not altered in the CTL group (P > 0.05).

Brachial Artery Endothelial Function

Scaled brachial FMD was similar between groups at baseline (P = 0.13), but group differences emerged over the 12-wk intervention (group × time interaction, P = 0.04; Fig. 1A). Scaled FMD increased with MICT from baseline to 6 wk (9.2 ± 2.4 to 11.4 ± 2.9%, P = 0.04), returning to baseline by 12 wk (9.5 ± 3.2%, P = 0.80 vs. baseline; P = 0.12 vs. 6 wk). In contrast, scaled brachial FMD did not change with SIT from baseline to 6 wk (7.0 ± 2.4 to 8.1 ± 2.4%, P = 0.25) or 12 wk (6.3 ± 2.4%, P = 0.43 vs. baseline), but there was a trend toward a reduction from 6 to 12 wk (−1.8 ± 2.5%, P = 0.06). No changes were observed in the CTL group from baseline to 6 wk (6.9 ± 2.9 to 6.0 ± 2.7%, P = 0.47) or 12 wk (5.8 ± 2.8%, P = 0.38 vs. baseline), or between 6 and 12 wk (P = 0.86). Between groups, MICT elicited a larger scaled FMD response than SIT and CTL at both 6 wk (two-way ANOVA, P = 0.003; individual group comparisons: MICT vs. SIT, P = 0.02; MICT vs. CTL, P = 0.001) and 12 wk (two-way ANOVA, P = 0.04; individual group comparisons: MICT vs. SIT, P = 0.03; MICT vs. CTL, P = 0.02). No training effects were observed for unscaled relative brachial FMD (P = 0.54), or any indexes of the brachial RH response: peak RH BF (P = 0.19), MBV to peak RH diameter (P = 0.30), SR AUC to peak RH diameter (P = 0.96), or time to peak RH diameter (P = 0.50) (Table 2).

Fig. 1.

Allometrically scaled brachial FMD (A) and resting brachial diameter (B) at baseline and 6 and 12 wk in MICT, SIT, and CTL groups. Group means and SDs are presented on the left, whereas individual data are shown on the right. Scaled FMD group means and SDs were back-transformed from linear mixed-model estimated means and SEs, respectively. Individual scaled FMD data were estimated using unstandardized β from linear regression analyses. Group × time interaction for scaled brachial %FMD (P = 0.04): *P < 0.05 vs. MICT baseline; †P < 0.05 vs. SIT and CTL at same time. Main effect of time for resting brachial diameter (P = 0.03, two-way ANOVA): *P < 0.05 vs. MICT baseline (follow-up analysis with one-way ANOVA).

Table 2.

Brachial and popliteal artery FMD characteristics

	MICT			SIT			CTL
	Baseline	6 Weeks	12 Weeks	Baseline	6 Weeks	12 Weeks	Baseline	6 Weeks	12 Weeks	P Value
Brachial artery (n = 24)
Resting diameter, mm	4.07 ± 0.22	4.28 ± 0.29	4.39 ± 0.33	4.13 ± 0.52	4.12 ± 0.46	4.15 ± 0.52	3.75 ± 0.51	3.83 ± 0.56	3.86 ± 0.42	0.14
Peak RH diameter, mm	4.45 ± 0.29	4.70 ± 0.28	4.73 ± 0.34	4.41 ± 0.53	4.44 ± 0.41	4.41 ± 0.56	4.03 ± 0.51	4.07 ± 0.56	4.11 ± 0.38	0.11
Absolute FMD, mm	0.38 ± 0.12	0.42 ± 015	0.34 ± 0.10	0.28 ± 0.10	0.32 ± 0.09	0.26 ± 0.07	0.28 ± 0.09	0.24 ± 0.07	0.25 ± 0.09	0.60
Relative FMD, %	9.3 ± 2.9	9.9 ± 3.9	7.9 ± 2.5	6.9 ± 3.0	8.0 ± 2.8	6.2 ± 1.3	6.3 ± 1.3	6.2 ± 1.7	6.7 ± 2.5	0.54
Resting BF, ml/min	57 ± 26	93 ± 56*	90 ± 35*†	48 ± 15	53 ± 27	51 ± 19	46 ± 30	49 ± 29	40 ± 22	0.04
Peak RH BF, ml/min	422 ± 121	419 ± 155	489 ± 182	364 ± 86	456 ± 154	442 ± 161	369 ± 210	365 ± 136	353 ± 171	0.19
MBV to peak, cm/s	30 ± 7	29 ± 10	33 ± 8	29 ± 7	32 ± 8	30 ± 5	31 ± 14	28 ± 9	27 ± 11	0.30
SR AUC to peak, ×103	30 ± 15	28 ± 16	31 ± 12	27 ± 14	27 ± 8	28 ± 9	29 ± 7	29 ± 7	27 ± 10	0.96
Time to peak, s	45 ± 10	50 ± 9	44 ± 11	43 ± 7	44 ± 11	46 ± 11	45 ± 7	50 ± 10	52 ± 9	0.50
Popliteal artery (n = 22)
Resting diameter, mm	6.58 ± 0.93	6.36 ± 0.65	6.51 ± 0.82	6.75 ± 1.08	6.24 ± 0.95	6.37 ± 1.12	5.80 ± 0.95	5.76 ± 1.13	5.86 ± 1.01	0.38
Peak RH diameter, mm	6.82 ± 0.92	6.63 ± 0.63	6.80 ± 0.85	6.99 ± 1.10	6.51 ± 1.00	6.54 ± 1.07	6.14 ± 0.99	6.04 ± 1.12	6.16 ± 1.03	0.27
Absolute FMD, mm	0.24 ± 0.27	0.27 ± 0.25	0.30 ± 0.23	0.24 ± 0.33	0.27 ± 0.25	0.17 ± 0.14	0.34 ± 0.12	0.29 ± 0.13	0.29 ± 0.18	0.79
Relative FMD, %	3.8 ± 4.5	4.3 ± 4.1	4.6 ± 3.9	3.8 ± 4.8	4.4 ± 4.0	2.9 ± 2.7	6.0 ± 2.5	5.2 ± 2.7	5.1 ± 3.6	0.84
Resting BF, ml/min	75 ± 47	61 ± 32	83 ± 44	81 ± 43	71 ± 45	63 ± 50	48 ± 35	52 ± 36	53 ± 27	0.23
Peak RH BF, ml/min	734 ± 131	587 ± 153	737 ± 182	796 ± 240	682 ± 326	693 ± 260	565 ± 311	519 ± 326	541 ± 252	0.14
MBV to peak, cm/s	18 ± 6	10 ± 5	16 ± 6	20 ± 10	18 ± 11	13 ± 6	15 ± 5	16 ± 5	10 ± 4	0.19
SR AUC to peak, ×103	14 ± 11	13 ± 7	14 ± 6	13 ± 5	15 ± 8	15 ± 8	13 ± 4	11 ± 6	13 ± 3	0.86
Time to peak, s	60 ± 33	107 ± 51	88 ± 48	64 ± 37	87 ± 55	98 ± 38	77 ± 25	64 ± 44	102 ± 36	0.36

Values are means ± SD; n, no. of subjects. RH, reactive hyperemia; FMD, flow-mediated dilation; Resting BF, 30-s average blood flow at rest; Peak RH BF, peak postdeflation blood flow; MBV to peak, average postdeflation mean blood velocity up to peak RH diameter; SR AUC to peak, shear rate area under curve to peak RH diameter; Time to peak, time to peak RH diameter. P values are group × time interaction. Main effect of time for brachial resting diameter (P = 0.03, baseline < 6 and 12 wk) and brachial peak RH diameter (P = 0.03, baseline < 6 and 12 wk) are shown. Main effect of group for absolute FMD (P = 0.01, MICT > SIT and CTL) is shown.

^*P < 0.001 vs. MICT baseline.

^†P < 0.01 vs. SIT and CTL at 12 wk.

Brachial Artery Diameter and Flow

Resting brachial artery diameter was similar between groups at baseline (P = 0.23) and increased over 12 wk (main effect of time, P = 0.03; Fig. 1B). MICT elicited the largest relative increase in resting diameter from baseline to 12 wk, 8%, compared with 0.5% in SIT and 3% in CTL. Follow-up analyses confirmed significant increases in resting diameter with MICT (one-way ANOVA, P = 0.01; individual time comparisons: baseline vs. 6 wk, P = 0.002; baseline vs. 12 wk, P = 0.003; 6 wk vs. 12 wk, P = 0.33), but not SIT (P = 0.91) or CTL (P = 0.63). Concomitant increases were observed in resting brachial artery BF, which was similar between groups at baseline (P = 0.60), but changed only with MICT (group × time interaction, P = 0.04; Table 2). Resting BF increased with MICT from baseline to 6 wk (+36 ml/min, P < 0.001) and 12 wk (+33 ml/min, P < 0.001 vs. baseline), with no differences between 6 and 12 wk (P = 0.80). No changes were observed from baseline to 12 wk with SIT (+3 ml/min, P = 0.77) or CTL (−6 ml/min, P = 0.53). Peak brachial RH diameter was similar between groups at baseline (P = 0.17) and increased over 12 wk (main effect of time, P = 0.03; Table 2). MICT elicited the largest increases in peak RH diameter, 6%, compared with 0% in SIT and 2% in CTL. Follow-up analyses confirmed significant increases in peak RH diameter with MICT (one-way ANOVA, P = 0.004; individual time comparisons: baseline vs. 6 wk, P < 0.001; baseline vs. 12 wk, P = 0.004; 6 wk vs. 12 wk, P = 0.70), but not SIT (P = 0.86) or CTL (P = 0.82).

Popliteal Artery Endothelial Function and Diameter

Training effects were not observed for any of the popliteal artery measures: absolute FMD (P = 0.79), relative FMD (P = 0.84), resting BF (P = 0.23), peak RH BF (P = 0.14), MBV to peak RH diameter (P = 0.19), SR AUC to peak RH diameter (P = 0.86), time to peak RH diameter (P = 0.36), resting diameter (P = 0.38), or peak RH diameter (P = 0.27) (Table 2). We did not allometrically scale popliteal FMD, since popliteal diameter did not differ between groups at baseline (P = 0.21) or change over the 12 wk.

Arterial Stiffness and Resting Hemodynamics

No interactions or main effects were observed for central arterial stiffness (carotid distensibility, P = 0.50; central PWV, P = 0.28) or lower limb arterial stiffness (leg PWV, P = 0.52) (Table 3). Similarly, resting systolic blood pressure (P = 0.31), diastolic blood pressure (P = 0.26), mean arterial pressure (P = 0.12), and HR (P = 0.85) did not change with training (Table 3).

Table 3.

Arterial stiffness and resting hemodynamics

	MICT			SIT			CTL
Variable	Baseline	6 Weeks	12 Weeks	Baseline	6 Weeks	12 Weeks	Baseline	6 Weeks	12 Weeks	P Value
Distensibility, ×10−3 mmHg−1	4.6 ± 0.7	4.8 ± 1.0	4.5 ± 1.2	4.6 ± 1.0	4.3 ± 1.0	4.9 ± 1.3	5.3 ± 1.3	5.0 ± 0.6	5.2 ± 1.4	0.50
Central PWV, m/s	6.8 ± 0.7	6.5 ± 0.8	6.6 ± 1.1	6.6 ± 0.9	6.7 ± 0.8	6.7 ± 0.5	6.5 ± 0.6	6.9 ± 0.8	6.4 ± 0.6	0.28
Leg PWV, m/s	7.1 ± 1.2	7.0 ± 1.1	6.9 ± 1.2	6.8 ± 1.0	7.0 ± 1.1	7.2 ± 0.8	6.7 ± 0.7	7.6 ± 1.3	7.5 ± 0.6	0.52
SBP, mmHg	112 ± 8	107 ± 9	111 ± 9	116 ± 8	111 ± 10	112 ± 8	110 ± 13	109 ± 11	109 ± 11	0.31
DBP, mmHg	67 ± 5	65 ± 6	66 ± 5	68 ± 3	67 ± 5	68 ± 7	67 ± 6	70 ± 9	69 ± 8	0.26
MAP, mmHg	85 ± 4	83 ± 5	84 ± 5	87 ± 3	85 ± 5	86 ± 5	84 ± 7	86 ± 6	86 ± 8	0.12
HR, beats/min	64 ± 11	61 ± 11	63 ± 8	64 ± 10	62 ± 10	62 ± 9	61 ± 12	60 ± 9	63 ± 10	0.85

Values are means ± SD. PWV, pulse wave velocity (central: carotid to femoral, leg: femoral to dorsalis pedis); SBP, systolic blood pressure; DBP, diastolic blood pressure; MAP, mean arterial pressure; HR, heart rate. P values are group × time interaction. No interactions or main effects were observed.

DISCUSSION

This is the first study to comprehensively compare the effects of MICT and SIT in both nonactive and active peripheral arteries in sedentary, healthy men. MICT induced changes in the brachial artery that were not observed with SIT, while neither training program induced changes in the active popliteal artery. These findings suggest that brachial artery responses to the intense, but brief and intermittent, nature of SIT may follow a different time course not captured by our 6- and 12-wk assessments, or may not occur at all.

Brachial Artery

In the brachial artery, MICT elicited changes in endothelial function and arterial diameter that were not mirrored with SIT. Allometric scaling unmasked group differences in endothelial function that were not detected in the analysis of unscaled relative FMD. The 2.2% increase in scaled FMD observed with MICT is comparable to the 1.7–3.2% increases previously reported in healthy men by studies that used protocols of similar intensity and duration (5, 8, 36). With no accompanying changes in RH BF, the increased scaled FMD observed with MICT is likely not the result of an augmented shear stimulus, but rather a reflection of training-induced improvements in brachial endothelial function. The greatest increases in brachial FMD have been reported after 2 wk of endurance training, with a gradual return to baseline levels after 4−6 wk (5, 36). Our 6-wk assessment is at the end of the previously observed window for capturing endothelial function responses in young healthy individuals; consequently, we may have missed peak FMD improvements. Interestingly, our MICT group had a baseline relative brachial FMD of 9.3%, which is larger than previously reported baseline (2.2, 5.8, and 5.9%), and training-induced peak FMD responses (3.9, 8.6, and 9.1%) (5, 8, 36). As such, it is possible functional responses to MICT progressed slower or lingered longer, enabling us to still see significant increases at our 6-wk assessment. Given the trend for reductions in scaled brachial FMD in our SIT group from 6 to 12 wk, it is unclear whether SIT may have elicited increases in brachial FMD before the 6-wk assessment, whether it would have elicited significant reductions if training were extended past the 12-wk assessment, or whether it simply does not alter brachial FMD. The direction and time course of the brachial FMD response to SIT in sedentary, healthy adults warrants further investigation.

Mechanistically, peripheral artery function changes when the endothelium repeatedly experiences elevated levels of shear stress (16). While it is technically challenging to measure shear stress during exercise, our findings suggest that MICT might provide a greater shear stimulus compared with low-volume SIT. The lack of response we observed in brachial artery endothelial function with SIT is in contrast to previous reports of interval training eliciting comparable, or superior, responses to MICT (27). This discrepancy may be the result of differences in the populations studied and the interval training models utilized. In the meta-analysis by Ramos et al. (27), interval training induced superior endothelial function changes compared with MICT in persons with elevated CVD risk (postinfarction, coronary artery disease, hypertension, metabolic syndrome, obesity, type 2 diabetes, and postmenopausal women), and presumably with impaired baseline endothelial function. Exercise training is known to elicit greater functional improvements in persons with cardiometabolic disorders (2); therefore, interval training may be a potent stimulus in impaired, but not healthy, vasculature. Additionally, interval training models previously used in clinical populations have had work intervals that were longer in duration and lower in intensity (i.e., 4 × 4 min at 90–95% peak HR; 10 × 1 min at 80–104% peak power output; 4–6 × 1 min at 80–85% V̇o_2peak) compared with the model used in the present study. Therefore, it may also be that our 3 × 20-s SIT model was too brief to induce a sufficient shear stimulus, and thus functional changes, in persons without baseline impairments in endothelial function.

Despite a main effect of time in resting brachial diameter, the greatest increases were observed in MICT, whereas changes in SIT and CTL were small and within our laboratory’s established day-to-day variability (3% coefficient of variance; unpublished data). Our findings are in agreement with Sawyer et al. (28), who recently reported a similar 4.9% increase in brachial artery diameter with 8 wk of a comparable MICT protocol, but no change in diameter with 10 × 1-min high-intensity interval training in obese adults. No changes in resting brachial diameter have been previously reported with lower limb training in sedentary, healthy men (8, 10, 33, 36); however, increases in vasodilator capacity were observed after an 8-wk combined cycling and treadmill program (36). Vasodilator capacity may be a more robust index of arterial remodeling, as resting diameter reflects changes in structure and/or vascular tone (34). Nevertheless, blood pressure and HR remained steady across visits, suggesting sympathetic tone was unaltered by training, and that the MICT-induced increase in resting brachial diameter likely reflects structural remodeling. This is further supported by the concomitant increase we observed in resting brachial BF with MICT, as a larger conduit artery diameter is able to facilitate more BF through the artery. Contrary to our hypothesis, SIT did not elicit any changes in resting diameter. SIT-induced structural remodeling may occur at a much slower rate than with MICT, or the brief and intermittent nature of low-volume SIT may not be sufficient to induce structural changes in the brachial artery.

Popliteal Artery

An unexpected finding in this study was that neither MICT nor SIT elicited changes in the active popliteal artery. Our laboratory previously showed that 6 wk of MICT and traditional Wingate-based SIT similarly improved relative FMD in the popliteal artery, without changing arterial diameter (26). Others have shown training-induced arterial remodeling after 4−8 wk in the popliteal artery (36), and after 8 and 12 wk in the femoral artery (10, 33). Thus we anticipated that low-volume SIT involving 1 min of intense intermittent exercise would elicit comparable responses to MICT, and that extending training to 12 wk would enable us to observe changes in arterial diameter. Compared with previous studies that successfully induced training responses, our participants had a larger resting popliteal diameter (6.4 mm vs. 5.6 mm and 4.9 mm) (26, 36). The popliteal artery is not uniform throughout its length, and normative values suggest we imaged the proximal (6.9 ± 0.9 mm) or midlevel (6.8 ± 0.8 mm), whereas previous studies imaged the smaller distal level (4.9 ± 0.6 mm) (39). The mechanisms by which the heterogeneity in size, and perhaps structural morphology along the artery, impact endothelial function are poorly understood. Nevertheless, popliteal diameter and relative FMD are negatively and moderately correlated (r = −0.48, P = 0.03) (32). This may elucidate why, compared with other studies, relative FMD in our training participants was smaller at baseline (3.8 vs. 5.0 and 6.2%) and did not change with training (±0.9 vs. +1.7 and +3.3%) (26, 36). These findings suggest that baseline popliteal diameter not only influences the FMD response, but it may also impact the magnitude of training effects in the FMD response. With limited data and discrepancies between studies, more research is required to determine the degree to which training responses depend on baseline diameter and/or the level along the popliteal artery, and whether responses differ between MICT and SIT.

Arterial Stiffness

Traditional endurance training has been shown to improve various indexes of stiffness in sedentary, healthy adults (9, 17, 20, 26, 30). Fewer studies have investigated the effects of SIT (or high-intensity interval training), and all have used protocols involving 20–40 min/session (7, 19, 26). Exercise training studies ranging from 6 days to 16 wk have reported significant reductions in central PWV of 0.3−0.5 m/s (7, 9, 19, 20). In the present study, despite comparable baseline values and a similar 0.3 m/s reduction after 6 wk of MICT, we did not observe any training effects on central PWV. Our ability to detect significant changes may have been limited by the interindividual variability in the change from baseline to 6 wk, which was twice as large as the mean difference (0.3 ± 0.6 m/s). Consistent with our central PWV results, and past findings (26), we did not observe changes in carotid artery distensibility, a localized index of central stiffness. Furthermore, 6 and 12 wk of MICT or low-volume SIT did not change leg PWV. In sedentary, healthy adults, previous observations of training effects on lower limb stiffness are limited and conflicting. Our findings are in agreement with Hayashi et al. (17) who reported no change in leg PWV after 16 wk of brisk walking/jogging. However, decreases in leg PWV were previously observed after 6 days of a 2-h cycling program (9), suggesting that changes in lower limb stiffness may require more intensive training stimuli. Training effects may also depend on baseline stiffness levels, as leg PWV was lower in the present study, 7.0 vs. 9.7 and 9.8 m/s (9, 17), suggesting that our participants already had reduced lower limb stiffness before commencing the training program. Overall, our findings suggest that 12 wk of traditional MICT or low-volume SIT did not induce changes in central or lower limb arterial stiffness in sedentary, healthy men.

Limitations and Strengths

This study assessed resting arterial diameter, which we acknowledge is neither a direct nor comprehensive indicator of arterial structure. Despite observing increases in resting and peak RH diameters in the brachial artery, these measures do not reflect an artery’s true peak dilatory capacity, and thus are not ideal surrogates for examining arterial remodeling. Future studies assessing structural responses to exercise training are encouraged to induce peak arterial dilation using a combination of nitroglycerin administration, forearm ischemia, and ischemic exercise (24). As our laboratory previously observed changes in popliteal endothelial function, but not arterial diameter, following 6 wk of MICT and traditional Wingate-based SIT (26), the time points selected in the present study allow us to make comparisons to, and build on, our previous findings. However, earlier and more frequent assessments would have enabled us to capture peak functional changes with MICT and any changes elicited by SIT that may have been missed, as well as establish a time course for potential brachial artery responses to low-volume SIT. Additionally, since we observed MICT-induced changes in endothelial function and diameter in the brachial artery, it would have been interesting to assess arterial stiffness in the nonactive upper limb (i.e., carotid-radial PWV). A major strength of this study is our comprehensive comparison of the effects of MICT and low-volume SIT on the arteries of healthy, untrained adults, as well as our inclusion of a nontraining control group.

Conclusions

We demonstrated that 6 and 12 wk of traditional MICT were superior to low-volume SIT at eliciting changes in the nonactive brachial artery. Despite resulting in similar improvements in aerobic capacity, MICT induced increases in brachial endothelial function and arterial diameter that were not apparent with low-volume SIT involving a total of 1 min of “all out” exercise per session. Arterial responses to the intense, but brief and intermittent, SIT stimulus may follow a different time course not captured with our 6- and 12-wk visits, or may not occur at all. Future studies should incorporate earlier and more frequent assessments to determine the time course of potential arterial responses to low-volume SIT. Investigations of other interval training models may also elucidate the specific characteristics of stimuli required to elicit functional and structural arterial responses in healthy adults.

GRANTS

NSERC Discovery and Research Tools and Instruments grants to M. J. MacDonald and M. J. Gibala funded equipment and software used in this study.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

N.S., J.B.G., M.J.G., and M.J.M. conceived and designed research; N.S. performed experiments; N.S. analyzed data; N.S. and M.J.M. interpreted results of experiments; N.S. prepared figures; N.S. drafted manuscript; N.S., J.B.G., M.J.G., and M.J.M. edited and revised manuscript; N.S., J.B.G., M.J.G., and M.J.M. approved final version of manuscript.