The impact of sprint interval training versus moderate intensity continuous training on blood pressure and cardiorespiratory health in adults: a systematic review and meta-analysis

Search strategy

The search strategy was designed by two authors (WL and SW) by an initial scoping review of the literature. Disagreement was handled by discussion with third author (XY). We conducted a systematic search of various electronic databases, including PubMed, Embase, and Web of Science, from database inception to July 2023.

The Boolean operators “OR” and “AND” were used, and the search was conducted using Mesh terms along with their respective synonyms. The Boolean search syntax displayed below was applied: “sprint interval training” OR “sprint interval exercise” OR “Sprint intermittent training” OR “sprint training” OR “sprint-interval training” OR “High-Intensity Interval Training*” OR “High intensity intermittent training” OR “High-intensity intermittent training” OR “Interval training” OR “interval exercise” OR “HIIT” OR “high intensity exercise” OR “high intensity aerobic interval training” OR “all-out exercise training” OR “all-out training” OR “all-out interval training” OR “Supramaximal interval training” OR “Wingate training” AND “Blood Pressure*” OR “Post-Exercise Hypotension” OR “DBP” OR “SBP” OR “Hypotension” OR “hypertension”.

Details of the literature search strategy are available in Supplementary Material. Qualification screening of the identified studies’ titles and abstracts was independently conducted by WL and CL based on the aforementioned keywords. The full text of studies that met the inclusion criteria was retrieved. Disagreements were resolved by consensus or consultation with XY.

Study eligibility criteria

Screened studies had to meet the following inclusion criteria: (1) Randomized trials published in English; (2) analyzed human participants of both sexes, aged >18 years; (3) included SIT protocols; (4) compare the effects of SIT with moderate-intensity continuous training; (5) resting BP was the primary outcomes and VO₂max was secondary outcomes; (6) intervention duration ≥2 weeks.

Studies with interventions mixing SIT with other interventions (e.g., dietary interventions, resistance training, hypoxia) and participants with a history of cardiovascular disease or other chronic conditions affecting blood pressure were excluded from the current review excluding individuals with hypertension.

Given the review’s objective, it is crucial to differentiate between MICT, SIT, and HIIT.

HIIT intensity ranges between 80–100% of maximum heart rate (HRmax) or maximum oxygen consumption (VO₂max), while SIT intensity generally exceeds 100% HRmax/VO₂max (e.g., 30 s of all-out exercise).The intensity of SIT typically exceeds 100% of VO₂max, with common exercise protocols involving 4-6 sets of 30 s of maximal effort exercise (Gist et al., 2014; Naves et al., 2018). MICT, on the other hand, generally ranges between 46–64% of VO₂max/HRmax or 40%–60% HR reserve/VO₂ reserve (Pescatello & Medicine AC of S, 2014).

Data extraction

All retrieved articles were imported into Endnote X9, and duplicate articles were removed. Preliminary screening of articles was conducted by reviewing titles and abstracts, and the full texts of potentially eligible studies were downloaded and read for further assessment. Full-text studies that did not meet the inclusion criteria were excluded and reasons for their exclusion documented (Fig. 1). Data extraction from the included studies was performed using a pre-designed table, capturing the following information: first author’s name, BP category, AGE, sample size, Training mode, Duration, Frequency, SIT protocol, MICT protocol. The study selection and data extraction processes were independently carried out by two reviewers (WL and CL), with mutual cross-verification. Any discrepancies were resolved through consultation with a third reviewer (XY). In cases where direct data were unavailable, authors of included studies were contacted.

Quality assessment and risk of bias

Two reviewers (GY and WL) independently evaluated studies that met the inclusion criteria for quality, using Review Manager 5.4 software (Cochrane Collaboration, London, UK). The tool was slightly modified to suit the study design and consisted of the following items: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting (for randomized controlled trials), and other biases. To evaluate the asymmetry of publication bias, effect sizes and standard errors were used to create funnel plots in the RevMan software.

Statistical analyses

Statistical analyses were carried out using Review Manager 5.4 software (Informer Technologies, Inc., Los Angeles, CA, USA). Changes in the mean and standard deviation of outcome measures were used to conduct between-group and within-group meta-analyses. Change in post-intervention mean was calculated by subtracting baseline from post-intervention values. Change in the SD of post-intervention outcomes was calculated using the Review Manager. Effect sizes (ES) were measured using mean, SD, and sample size. Summary estimates with 95% confidence intervals were pooled using the DerSimonian-Laird random effects model or fixed effects model according to between-study heterogeneity (DerSimonian & Laird, 1986).

The I² statistic, expressed as a percentage, was used to determine heterogeneity between studies, where an I² ≥ 75% implied high heterogeneity, 75% >I² ≥ 50% implied moderate heterogeneity, and I² < 50% indicated low heterogeneity (Higgins et al., 2003). Forest plots were used to illustrate summary statistics and the variation across studies.

Subgroup analysis was conducted to analyze whether population and training characteristics in the study affected blood pressure. Duration and single sprint time were examined to identify any heterogeneity. Funnel plots were used to assess publication bias. Due to the limited number of included studies, univariate meta-regression analyses were not conducted.

Study selection

A PRISMA diagram of literature search and selection was presented in Fig. 1. The initial search yielded 1874 articles, 1476 of which were considered relevant after eliminating duplicates. Subsequently, 1068 studies were excluded based on assessments of their titles and abstracts. The eligibility of the remaining 408 full-text articles was scrutinized, resulting in the exclusion of 400 articles for various reasons. As a result, eight articles were included in the final meta-analysis.

Characteristics of the studies

Table 1 presents an overview of participant characteristics. In total, 169 participants were analyzed across all studies: 84 involved in SIT and 85 in MICT. Among the included studies, three (Rakobowchuk et al., 2008; Boer et al., 2014; Cuddy, Ramos & Dalleck, 2019) analyzed adults of both sexes (37.5%), four (Skleryk et al., 2013; Cocks et al., 2013; Shenouda et al., 2017; Petrick et al., 2021) exclusively analyzed male participants (50.0%), and one (Cocks et al., 2016) exclusively analyzed female participants (12.5%).

Of the eight studies, seven (Rakobowchuk et al., 2008; Skleryk et al., 2013; Cocks et al., 2013; Cocks et al., 2016; Shenouda et al., 2017; Cuddy, Ramos & Dalleck, 2019; Petrick et al., 2021) included untrained or sedentary participants (87.5%), while one (Boer et al., 2014) did not report this information (12.5%). The participants’ average age ranged from 18 to 52 years, and their BMI values ranged between 23.3 and 34.4 kg/m². Five studies (Rakobowchuk et al., 2008; Cocks et al., 2013; Boer et al., 2014; Shenouda et al., 2017; Petrick et al., 2021) enrolled participants with normal blood pressure, while the remaining three (Skleryk et al., 2013; Cocks et al., 2016; Cuddy, Ramos & Dalleck, 2019) included participants with pre-hypertension or hypertension. Only one (Petrick et al., 2021) study involved participants who were taking anti-hypertensive drugs.

Table 1 presents an overview of SIT and MICT interventions used in the included studies. MICT sessions ranged from 40 to 65% of VO₂peak or reserve heart rate. All four MICT studies (Rakobowchuk et al., 2008; Cocks et al., 2013; Cocks et al., 2016; Shenouda et al., 2017) lasted 40 to 60 min, while the remaining four MICT studies (Skleryk et al., 2013; Boer et al., 2014; Cuddy, Ramos & Dalleck, 2019; Petrick et al., 2021) lasted from 25 to 40 min. The SIT protocols varied in duration, frequency, and intensity across studies. All studies employed a cycling exercise that involved four (Rakobowchuk et al., 2008; Cocks et al., 2013; Cocks et al., 2016; Petrick et al., 2021) to six 30-second bouts of maximal cycling efforts, followed by a 3 to 4 min rest between bouts. Intervention durations spanned from 2 to 12 weeks, and SIT was performed three times per week in most studies, whereas MICT was performed five times per week (Skleryk et al., 2013; Cocks et al., 2013; Cocks et al., 2016; Eskelinen et al., 2016; Shenouda et al., 2017; Petrick et al., 2021) .

All studies did not mention the implementation of warm-up and cool-down protocols for MICT. Regarding SIT protocols, four studies (Boer et al., 2014; Shenouda et al., 2017; Cuddy, Ramos & Dalleck, 2019; Petrick et al., 2021) reported warm-up and cool-down durations of 2-5 min; two studies (Skleryk et al., 2013; Cocks et al., 2013) mentioned a warm-up protocol but did not specify a cool-down protocol; and two studies (Rakobowchuk et al., 2008; Cocks et al., 2016) did not provide information on either warm-up or cool-down protocols.

All studies reported blood pressure as an outcome evaluation, with Table 2 summarizing the values of SBP and DBP before and after the intervention for each study.

Risk of bias within and across studies

Risk of bias was evaluated for the eight studies, with most articles showing low or unclear risk in key areas. Low risk was found in a high percentage of studies for incomplete outcome data, selective reporting, random sequence, and other biases, while a moderate percentage was found in allocation concealment. However, blinding of participants and personnel, as well as outcome assessment, showed a high percentage of unclear risk across the studies. Only one study (Boer et al., 2014) reported blinded assessors quantifying all checked variables. Additionally, a few studies (Rakobowchuk et al., 2008; Petrick et al., 2021) showed a high risk of bias in the generation of random sequence and allocation concealment, as seen in Fig. 2.

Meta-analysis

Within-group effects of blood pressure

SIT and MICT interventions’ effect on SBP and DBP levels was evaluated before and after the intervention in eight studies.

The meta analysis revealed a significant decrease in SBP (MD = −2.82 mmHg, 95% CI [−4.53 to −1.10], p = 0.08, I² = 45%) in the SIT group after the intervention, with no significant changes observed in DBP (MD = −0.75 mmHg, 95% CI [−1.92 to 0.42], p = 0.16, I² = 33%). The heterogeneity was low for both analyses (I²). Forest plots of changes in resting SBP and DBP before and after intervention in the SIT group are shown in Fig. 3. Funnel plots showed no indication of publication bias (Fig. S1).

Compared to baseline values, the MICT group showed a significant decrease in both SBP (MD = −3.00 mmHg, 95% CI [−5.31 to −0.69], p = 0.68, I² = 0%) and DBP (MD = −2.11 mmHg, 95% CI [−3.63 to −0.60], p = 0.72, I² = 0%) levels after the intervention, with low heterogeneity observed in both analyses. Forest plots of changes in resting SBP and DBP before and after intervention in the SIT group are shown in Fig. 4. Funnel plots showed no indication of publication bias (Fig. S2).

Between-group effects of blood pressure

No significant differences were observed in resting SBP (MD = −0.92 mmHg, 95% CI [ −4.48 to 2.64], p = 0.61) changes between SIT and MICT interventions from pre- to post-intervention. However, moderate heterogeneity was detected for this analysis (I² = 57%; p = 0.02). Similarly, no significant differences were found in resting DBP (MD = 1.34 mmHg, 95% CI [ −0.07 to 2.75], p = 0.06) between SIT and MICT interventions before and after the intervention, with lower heterogeneity observed (I² = 0%; p = 0.67). The forest plots for changes in resting SBP and DBP are shown in Fig. 5. Funnel plots showed no indication of publication bias (Fig. S3).

Cardiorespiratory fitness

SIT and MICT interventions’ effect on VO₂peak levels was evaluated in six studies. The meta-analysis demonstrated that both SIT and MICT significantly improved VO₂peak (SIT, MD = 1.75mL/kg/min, 95% CI [0.39–3.10], p = 0.02, I² = 61%; MICT, MD = 3.10mL/kg/min, 95% CI [1.03–5.18], p = 0.007, I² = 69%). However, the pooled results of the meta-analysis suggested that MICT was more effective in enhancing VO₂peak than SIT (MD = −1.36 mL/kg/min, 95% CI [−2.31 to 0.40], p = 0.56, I² = 0%). Forest plots for changes in VO₂peak are shown in Fig. 6.

Subgroup analysis

Considering that the duration of SIT and the single sprint time may trigger blood pressure changes of different magnitudes, we performed a subgroup analysis.

SIT versus MICT duration ≥8 weeks

Three studies (Boer et al., 2014; Shenouda et al., 2017; Cuddy, Ramos & Dalleck, 2019) investigated the effect of SIT duration greater than or equal to 8 weeks compared to MICT on SBP and DBP. The subgroup analysis of these studies indicated that when the duration was greater than or equal to 8 weeks, the SIT group showed a significant decrease in SBP (MD = −6.10 mmHg, 95% CI [ −10.51 to −1.70], p = 0.49, I² = 0%) and no significant change in DBP (MD = 0.47 mmHg, 95% CI [−2.41 to 3.36], p = 0.85, I² = 0%) compared with the MICT group. No heterogeneity was evident, and the forest plots of the effect of SIT duration ≥ 8 weeks versus MICT on SBP and DBP are shown in Fig. 7.

SIT versus MICT duration <8 weeks

Five studies (Rakobowchuk et al., 2008; Skleryk et al., 2013; Cocks et al., 2013; Cocks et al., 2016; Petrick et al., 2021) investigated the effect of SIT duration less than 8 weeks compared to MICT on SBP and DBP. For durations less than 8 weeks, no significant changes were observed in SBP (MD = 1.71 mmHg, 95% CI [−0.71 to 4.41], p = 0.24, I² = 28%), but the MICT group showed a better reduction in DBP (MD = 1.61 mmHg, 95% CI [−0.01 to 3.23], p = 0.39, I² = 3%) than the SIT group. No heterogeneity was evident for this analysis. The forest plots of the effect of SIT duration <8 weeks versus MICT on SBP and DBP are shown in Fig. 7.

SIT versus MICT sprint time ≥30

Four studies (Rakobowchuk et al., 2008; Cocks et al., 2013; Cocks et al., 2016; Petrick et al., 2021) investigated the effect of SIT sprint time greater than or equal to 30 s compared to MICT on SBP and DBP. The subgroup analysis showed that when the sprint time was greater than or equal to 30 s, there was no significant difference between the two groups in reducing SBP (MD = 2.20 mmHg, 95% CI [−0.43 to 4.84], p = 0.20, I² = 36%). However, the MICT group showed a better reduction in DBP than the SIT group (MD = 2.11 mmHg, 95% CI [0.38–3.83], p = 0.69, I² = 0%). No heterogeneity was detected, and forest plots of the effect of SIT sprint time ≥30 s versus MICT on SBP and DBP are shown in Fig. 8.

SIT versus MICT sprint time <30s

Subgroup analysis of four studies (Skleryk et al., 2013; Boer et al., 2014; Shenouda et al., 2017; Cuddy, Ramos & Dalleck, 2019) investigated the effect of SIT sprint time less than 30 s versus MICT on SBP and DBP. For sprint times less than 30 s, SIT was significantly more effective in reducing SBP compared to MICT (MD = −4.39 mmHg, 95% CI [−7.98 to −0.80], p = 0.37, I² = 5%), while the effect on reducing DBP was similar between the two groups (MD = −0.21 mmHg, 95% CI [−2.67 to 2.24], p = 0.77, I² = 0%). No heterogeneity was detected, and forest plots of the effect of SIT sprint time <30 s versus MICT on SBP and DBP are shown in Fig. 8.

The subgroup analysis showed that the reduction in BP in the SIT group was significantly impacted by the duration or sprint time of SIT relative to the MICT group (Table 3). For durations greater than or equal to 8 weeks, SIT was more effective in lowering SBP, whereas for durations less than 8 weeks, MICT was more effective in lowering DBP. When the sprint time was greater than or equal to 30 s, MICT was more effective in lowering DBP, whereas SIT was more effective in lowering SBP when the sprint time was less than 30 s.