Excessive sedentary behavior is a risk factor for arterial stiffness, cardiovascular disease, and mortality [1, 2]. Experimental studies report vascular impairment from prolonged sitting protocols of 3–6 h in length [3, 4], and interrupting sedentary behavior appears to mitigate these detrimental effects [3]. However, the acute effects of prolonged sitting on pulse wave velocity (PWV) are unclear. Guidelines recommend assessment of PWV in a supine posture [5], yet the true effect of prolonged sitting on PWV may be impacted by the seated-to-supine posture change prior to measurement. Though measurement in a seated posture may thus be preferred, the measurement properties of PWV while maintaining a seated posture are unknown. Thus, we studied the validity (two tests) and reliability (one test) of PWV assessment in a seated posture.

Twenty men and women with no contraindications for PWV testing participated in a randomized crossover study (SIT vs. SUPINE). The experimental conditions were separated by at least 2, but not more than 13, days and followed 12 h of abstention from food, caffeine, nicotine, alcohol, and moderate-to-vigorous physical activity. During each experimental visit, participants completed counterbalanced assessments of both seated and supine carotid-radial (crPWV) and carotid-ankle (caPWV) PWV after a 10-min rest. Following this, at each visit, in counterbalanced postures, subjects consumed 300 mg of caffeine, which served as a standardized perturbation. PWV was measured at 10, 20, and 30 min postcaffeine consumption. PWV was measured by a Complior Analyze system (ALAM Medical, France), with three separate scans of ten valid waveforms at each time point. Intraclass correlations (ICCs) were calculated to determine reliability across testing days; ICC ≥ 0.75 was considered good reliability. Pearson’s correlations (r) determined the validity of seated (alternative method) versus supine (gold standard) peripheral PWV measurements. Generalized linear mixed models evaluated differences in PWV response across time points, adjusting for age, sex, and condition order.

On average, participants were 25.9 (7.7) years old with a BMI of 25.5 (4.3) kg/m2 and resting blood pressure of 115/70 (10/9) mmHg. Half (50%) of the participants were male and 85% were white.

When measured in the supine posture (gold standard), day-to-day reliability had an ICC = 0.636 (95% Confidence Interval (CI) 0.279–0.838) for crPWV and an ICC = 0.361 (95% CI 0.099–0.694) for caPWV. In the seated posture (alternative method), ICC = 0.768 (95% CI 0.502–0.901) for crPWV, and ICC = 0.759 (95% CI 0.475–0.900) for caPWV. Only the seated peripheral PWV assessment met the criteria for good reliability (≥0.75) (Fig. 1).

Fig. 1
figure 1

Reliability of resting peripheral PWV measures. Mean ± SD calculated using an average of measurements within subjects across both days

Full size image

Supine versus seated peripheral PWV measurements at rest (Fig. 2a, b) suggested high agreement for crPWV (r = 0.812 95%, CI 0.563–0.943) but low agreement for caPWV (r = 0.277, 95% CI 0.119–0.543).

Fig. 2
figure 2

Validity of seated vs. supine PWV at rest and in response to caffeine. a, b Comparison of PWV measures between postures. c, d Response over time of PWV measures to a standard perturbation

Full size image

During the caffeine perturbation (Fig. 2c, d), peripheral PWV condition × time interactions were nonsignificant (p ≥ 0.10), suggesting similar responses across postures. Peripheral PWV was significantly higher in the seated vs. supine condition for both crPWV (βposture = 0.548 m/s, p < 0.001) and caPWV (βposture = 3.583 m/s, p < 0.001). While there was no time effect for crPWV (βtime = 0.009 m/s per 10 min, p = 0.198), caPWV increased over time in response to caffeine (βtime = 0.014 m/s per 10 min, p = 0.014).

These data suggest acceptable validity and reliability for measuring crPWV and possibly caPWV during prolonged sitting without interrupting subject posture. Seated peripheral PWV was found to have good day-to-day reliability. In fact, its reliability was better than that of the standard supine measurement. This could be due to the larger ranges in seated versus supine PWV, which may have enhanced our ability to observe relationships [6]. Furthermore, validity for seated PWV was demonstrated for crPWV at rest and in response to a caffeine perturbation, while caPWV demonstrated validity only during the caffeine response. The poor validity for resting caPWV could reflect the low reliability and range of resting supine caPWV, indicating a need for further research.

We know of only one other study examining the measurement properties of seated peripheral PWV, and it used cuff oscillometry [7]. In contrast to our results, the authors determined that seated PWV using cuff oscillometry did not meet the criteria for validity or reliability. It should be noted that the use of a different device, which differs in terms of the assessment technique as well as the algorithm used, could explain these disparate results.

While both flow-mediated dilation and, to a lesser extent, PWV have been used to research cardiovascular responses to prolonged sitting [3, 4, 8], PWV may be preferred as it is less technically demanding than other measurement techniques [5]. However, previous studies measuring PWV have done so in the recommended supine posture, which could have affected the results [3, 4, 8]. Although our results are promising for future research on acute cardiovascular responses (i.e., PWV) to prolonged sitting that may wish to maintain a seated posture, our study is not without limitations and should be interpreted with caution. Not being able to measure aortic PWV in the seated posture (since the femoral artery could not be palpated) limited our study to peripheral PWV. In addition, the use of a young, healthy population limits the generalizability of our findings. However, the strengths of our study include the simultaneous evaluation of both the day-to-day reliability and validity (at rest and in response to a perturbation) of seated versus supine peripheral PWV at two anatomical sites.

Future research should evaluate the measurement properties of seated versus supine aortic PWV and use more clinical populations. However, our data indicate that studies evaluating the impact of prolonged sitting and the breaking up prolonged sitting can and should consider measuring peripheral PWV in a seated posture, especially crPWV.