Aortic pulse wave velocity (PWV) which relates to vascular stiffness, independently predicts cardiovascular risk including stroke (Hametner et al., 2013; Vishram-Nielsen et al., 2020). PWV values above 10 m/s have been suggested as threshold indicating increased risk (Van Bortel et al., 2012) and are included in current hypertension guidelines for assessing hypertension-mediated organ damage and guiding secondary prophylaxis (Williams et al., 2018). Local PWV measurements are also feasible using two points close or more distant to each other as well as measurements of pressure-velocity loops at one single spot. However, the guidelines do not specify certain methods to measure PWV. Measurements at different sites including the thigh were used in large scale studies. However, this PWV assessment is difficult to implement in busy clinics or during daily routine. Oscillometric blood pressure monitors estimating PWV based on age and blood pressure could simplify vascular assessment (Hametner et al., 2013; Hametner et al., 2021). Indeed, estimated PWV correlates with age (Schwartz et al., 2019) and invasively measured PWV (Hametner et al., 2013). These monitors operate observer-independent, are easy to use, and are, therefore, applied in clinical practice and in clinical studies alike (Paiva et al., 2020). However, in a study investigating beneficial effects of competitive lifelong physical exercise in elite Masters Athletes, we obtained estimated PWV results suggesting that the approach adds little information to classical cardiovascular risk assessment. We decided comparing the methodology with aortic PWV measurements in an independent set of healthy persons participating in a head-down-tilt bedrest study.

During the 23rd Masters Athletics Championships 2018 in Málaga, Spain, we approached 163 master athletes. We excluded athletes with atrial fibrillation or significant cardiovascular disease assessed by echocardiography. In the remaining 129 athletes (88 men/41 women, 56 ± 11 [range 35–86] years, 24.0 ± 3.5 kg/m²), we acquired blood pressure, heart rate, and estimated PWV thrice on the same arm after 10 min supine rest (Arc-Solver-Algorithm, CardioCube, AIT, Vienna, Austria).

In 24 healthy participants (16 men/8 women, 33 ± 9 years, 24.3 ± 2.1 kg/m²) of the AGBRESA (artificial gravity bedrest) study, which was conducted in collaboration with NASA and ESA in Cologne, Germany (Hoffmann et al., 2021), we assessed estimated PWV before head-down-tilt bedrest as described above. As the estimated PWV has been validated against invasive catheter measurements derived from the ascending aorta to the aortic bifurcation (Hametner et al., 2013), we mimicked this approach non-invasively by measuring pulse wave arrival time from the electrocardiographic R-peak to the arrival of the pulse wave at an oscillometric thigh cuff. We corrected pulse wave arrival time for isovolumetric contraction time, the time from the R-peak in the ECG to aortic valve opening assessed by 2D-pulsed-wave-doppler echocardiography. Corrected pulse wave arrival time represents pulse wave travel time from the aortic valve opening to the arrival at the thigh. Moreover, in contrast to the carotid-femoral PWV, this approach includes the ascending aorta and the aortic arc, which are commonly affected by aging-associated vascular disease. Dividing jugulum-thigh cuff distance by corrected pulse wave arrival time resulted in measured PWV.

All subjects provided informed consent and consent to publish before enrollment. The bedrest study as well as the study in master athletes was approved by the Northrine-Medical-Association (Ärztekammer Nordrhein, 2018143 and 2018171) ethics committee and registered at the German Clinical Trial Register (DRKS00015677 and DRKS00015172).

According to Kolmogorov-Smirnov test, all data were normally distributed, and results are reported as mean values ± standard deviation. Based on other cardiovascular risk estimate models, we used a quadratic regression model.

Resting heart rate was 61 ± 11 bpm, and blood pressure was 128 ± 15/78 ± 8 mmHg in master athletes. Estimated PWV ranged from 5.5 to 14.5 with a mean of 8.3 ± 1.8 m/s. In a quadratic regression model, age and mean arterial pressure predicted estimated PWV (beta value 0.93 and 0.14, p<0.001), whereas, sex and BMI had no influence. The model explained 95% (R² = 0.95) of estimated PWV’s variance. Age alone explained 90% (R² = 0.90).

Resting heart rate was 62 ± 9 bpm, and blood pressure was 125 ± 11/70 ± 7 mmHg in the AGBRESA bedrest study. Estimated PWV was 5.8 ± 1.1 m/s, and measured PWV was 4.8 ± 0.6. Age and estimated PWV were highly correlated (R² 0.88, p<0.001, Figure 1A). The correlation between measured PWV and age was weaker (R² 0.55, p<0.001). In a quadratic regression model, age (beta value 0.99, p<0.001) but not BMI, sex, or mean arterial pressure predicted estimated PWV. The model explained 98% of estimated PWV’s variance. R² between estimated and measured PWV was 0.40 (p<0.05). We observed an increasing bias between estimated and measured PWV with advancing age with an average 0.96 ± 0.83 m/s difference between methods (Figure 1B). When adjusting the correlation of measured and estimated PWV for age we could no longer observe a significant relationship (p=0.267).

Figure 1

Download asset Open asset

To assess influences of entered rather than chronological age on estimated PWV, we repeated measurements in one male person, who is in his 50s and has normal BMI, and randomly changed the entered age in decades from 30 to 80 years. We obtained measurements in each entered age decade thrice. Again, estimated PWV related to entered age (R² 0.996, p<0.001, Figure 1C) akin to the relationship between estimated PWV and chronological age in athletes and in bedrest study participants.

The idea of obtaining vascular measurements, such as PWV, is gaining insight in individual vascular risk above and beyond traditional risk factors. The clinical goal is to target preventive measures to patients most likely to benefit. Similarly to other studies (Schwartz et al., 2019), we observed an almost perfect correlation between estimated PWV and age in two independent cohorts. Strikingly, estimated PWV in elite Master Athletes, which would be expected to benefit from lifelong exercise, and in healthy bedrest study participants of average physical fitness were located on the same regression line. Suffice it to say that many athletes were frustrated when we reported their findings.

Theoretically, our findings could result from methodological limitations of PWV estimates or a rather limited effect of lifelong exercise on aortic stiffness. A previous study showed significantly lower measured PWV in endurance trained compared with sedentary older men (Vaitkevicius et al., 1993) making a methodological limitation more likely. Indeed, in our study, estimated PWV substantially overestimated PWV, particularly in older persons. Moreover, the modest correlation between estimated and measured PWV was attenuated after adjusting for age in our relatively small bedrest cohort. Finally, in a previous study, estimated PWV highly correlated with PWV calculated solely from age and blood pressure (Hametner et al., 2021). Thus, the algorithm providing PWV estimates appears to weigh age so strongly that subtle influences on vascular wall properties caused by hemodynamic changes and physical training cannot be discerned. Even more so, estimated PWV mainly reflects data entered prior to measuring blood pressure rather than true vascular properties. The clinical implication is that estimated PWV is no substitute for measured PWV, which likely limits the utility of the methodology in individualizing risk assessment. In fact, simply asking the patient in front of us for his or her age and measuring blood pressure provides almost as much information as estimating PWV. Instead, vascular aging assessment using established methodologies rather than estimates should be considered (Jordan et al., 2015).

As we obtained personal health data from human subjects, we cannot make their raw data publicly available. However, an interested researcher is able to access the original data by sending a project proposal to the corresponding author. This project proposal will be reviewed by the medical board of the DLR Institute of Aerospace Medicine. If the project is scientifically valuable, the committee decides to what extent the original data can be made available. Commercial research is excluded from this option. The syntax used for statistical analysis and the numerical data used to generate the figures are stored in Dryad: https://doi.org/10.5061/dryad.8931zcrsb.

1. 1. Möstl S
   2. Hoffmann F
   3. Hönemann J
   4. Alvero-Cruz J
   5. Rittweger J
   6. Tank J
   7. Jordan J

   (2021) Dryad Digital Repository

   Estimated pulse wave velocity is of limited utility to assess vascular stiffness.

   https://doi.org/10.5061/dryad.8931zcrsb

1. 1. Hametner B
   2. Wassertheurer S
   3. Kropf J
   4. Mayer C
   5. Eber B
   6. Weber T

   (2013) Oscillometric estimation of aortic pulse wave velocity: comparison with intra-aortic catheter measurements

   Blood Pressure Monitoring 18:173–176.

   https://doi.org/10.1097/MBP.0b013e3283614168

   • PubMed
   • Google Scholar
2. 1. Hametner B
   2. Wassertheurer S
   3. Mayer CC
   4. Danninger K
   5. Binder RK
   6. Weber T

   (2021) Aortic Pulse Wave Velocity Predicts Cardiovascular Events and Mortality in Patients Undergoing Coronary Angiography: A Comparison of Invasive Measurements and Noninvasive Estimates

   Hypertension (Dallas, Tex) 77:571–581.

   https://doi.org/10.1161/HYPERTENSIONAHA.120.15336

   • PubMed
   • Google Scholar
3. 1. Hoffmann F
   2. Rabineau J
   3. Mehrkens D
   4. Gerlach DA
   5. Moestl S
   6. Johannes BW
   7. Caiani EG
   8. Migeotte PF
   9. Jordan J
   10. Tank J

   (2021) Cardiac adaptations to 60 day head-down-tilt bed rest deconditioning. Findings from the AGBRESA study

   ESC Heart Failure 8:729–744.

   https://doi.org/10.1002/ehf2.13103

   • PubMed
   • Google Scholar
4. 1. Jordan J
   2. Nilsson PM
   3. Kotsis V
   4. Olsen MH
   5. Grassi G
   6. Yumuk V
   7. Hauner H
   8. Zahorska-Markiewicz B
   9. Toplak H
   10. Engeli S
   11. Finer N

   (2015) Joint scientific statement of the European Association for the Study of Obesity and the European Society of Hypertension: Obesity and early vascular ageing

   Journal of Hypertension 33:425–434.

   https://doi.org/10.1097/HJH.0000000000000473

   • PubMed
   • Google Scholar
5. 1. Paiva AMG
   2. Mota-Gomes MA
   3. Brandão AA
   4. Silveira FS
   5. Silveira MS
   6. Okawa RTP
   7. Feitosa ADM
   8. Sposito AC
   9. Nadruz W

   (2020) Reference values of office central blood pressure, pulse wave velocity, and augmentation index recorded by means of the Mobil-O-Graph PWA monitor

   Hypertension Research 43:1239–1248.

   https://doi.org/10.1038/s41440-020-0490-5

   • PubMed
   • Google Scholar
6. 1. Schwartz JE
   2. Feig PU
   3. Izzo JL

   (2019) Pulse Wave Velocities Derived From Cuff Ambulatory Pulse Wave Analysis

   Hypertension (Dallas, Tex) 74:111–116.

   https://doi.org/10.1161/HYPERTENSIONAHA.119.12756

   • PubMed
   • Google Scholar
7. 1. Vaitkevicius PV
   2. Fleg JL
   3. Engel JH
   4. O’Connor FC
   5. Wright JG
   6. Lakatta LE
   7. Yin FC
   8. Lakatta EG

   (1993) Effects of age and aerobic capacity on arterial stiffness in healthy adults

   Circulation 88:1456–1462.

   https://doi.org/10.1161/01.cir.88.4.1456

   • PubMed
   • Google Scholar
8. 1. Van Bortel LM
   2. Laurent S
   3. Boutouyrie P
   4. Chowienczyk P
   5. Cruickshank JK
   6. De Backer T
   7. Filipovsky J
   8. Huybrechts S
   9. Mattace-Raso FUS
   10. Protogerou AD
   11. Schillaci G
   12. Segers P
   13. Vermeersch S
   14. Weber T
   15. Artery Society
   16. European Society of Hypertension Working Group on Vascular Structure and Function
   17. European Network for Noninvasive Investigation of Large Arteries

   (2012) Expert consensus document on the measurement of aortic stiffness in daily practice using carotid-femoral pulse wave velocity

   Journal of Hypertension 30:445–448.

   https://doi.org/10.1097/HJH.0b013e32834fa8b0

   • PubMed
   • Google Scholar
9. 1. Vishram-Nielsen JKK
   2. Laurent S
   3. Nilsson PM
   4. Linneberg A
   5. Sehested TSG
   6. Greve SV
   7. Pareek M
   8. Palmieri L
   9. Giampaoli S
   10. Donfrancesco C
   11. Kee F
   12. Mancia G
   13. Cesana G
   14. Veronesi G
   15. Kuulasmaa K
   16. Salomaa V
   17. Kontto J
   18. Palosaari T
   19. Sans S
   20. Ferrieres J
   21. Dallongeville J
   22. Söderberg S
   23. Moitry M
   24. Drygas W
   25. Tamosiunas A
   26. Peters A
   27. Brenner H
   28. Njolstad I
   29. Olsen MH
   30. On behalf of the MORGAM Project

   (2020) Does Estimated Pulse Wave Velocity Add Prognostic Information?

   Hypertension 75:1420–1428.

   https://doi.org/10.1161/HYPERTENSIONAHA.119.14088

   • Google Scholar
10. 1. Williams B
    2. Mancia G
    3. Spiering W
    4. Agabiti Rosei E
    5. Azizi M
    6. Burnier M
    7. Clement DL
    8. Coca A
    9. de Simone G
    10. Dominiczak A
    11. Kahan T
    12. Mahfoud F
    13. Redon J
    14. Ruilope L
    15. Zanchetti A
    16. Kerins M
    17. Kjeldsen SE
    18. Kreutz R
    19. Laurent S
    20. Lip GYH
    21. McManus R
    22. Narkiewicz K
    23. Ruschitzka F
    24. Schmieder RE
    25. Shlyakhto E
    26. Tsioufis C
    27. Aboyans V
    28. Desormais I

    (2018)

    esc/esh guidelines for the management of arterial hypertension: The task force for the management of arterial hypertension of the european society of cardiology and the european society of hypertension

    Journal of Hypertension 36:1953–2041.

    • PubMed
    • Google Scholar

Author details

1. Stefan Möstl

   Department of Cardiovascular Aerospace Medicine, Institute of Aerospace Medicine, German Aerospace Center, Cologne, Germany

   Contribution

   Data curation, Formal analysis, Investigation, Methodology, Writing - original draft

   Contributed equally with

   Fabian Hoffmann

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-4983-4754

2. Fabian Hoffmann

   1. Department of Cardiovascular Aerospace Medicine, Institute of Aerospace Medicine, German Aerospace Center, Cologne, Germany
   2. Department of Cardiology, University Hospital Cologne, Cologne, Germany

   Contribution

   Data curation, Formal analysis, Visualization, Writing - original draft

   Contributed equally with

   Stefan Möstl

   Competing interests

   No competing interests declared

3. Jan-Niklas Hönemann

   1. Department of Cardiovascular Aerospace Medicine, Institute of Aerospace Medicine, German Aerospace Center, Cologne, Germany
   2. Department of Cardiology, University Hospital Cologne, Cologne, Germany

   Contribution

   Visualization, Writing - original draft

   Competing interests

   No competing interests declared

4. Jose Ramon Alvero-Cruz

   Department of Human Physiology and Physical Sports Education, Faculty of Medicine, University of Málaga, Málaga, Spain

   Contribution

   Conceptualization, Project administration, Writing - review and editing

   Competing interests

   No competing interests declared

5. Jörn Rittweger

   Department of Muscle and Bone Metabolism, Institute of Aerospace Medicine, German Aerospace Center, Cologne, Germany

   Contribution

   Conceptualization, Project administration, Writing - review and editing

   Competing interests

   No competing interests declared

6. Jens Tank

   Department of Cardiovascular Aerospace Medicine, Institute of Aerospace Medicine, German Aerospace Center, Cologne, Germany

   Contribution

   Conceptualization, Project administration, Resources, Supervision, Writing - review and editing

   Competing interests

   No competing interests declared

7. Jens Jordan

   1. Institute of Aerospace Medicine, German Aerospace Center, Cologne, Germany
   2. Chair of Aerospace Medicine, University of Cologne, Cologne, Germany

   Contribution

   Resources, Supervision, Writing - original draft

   For correspondence

   Jens.Jordan@dlr.de

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-4518-0706

• 484

  views

• 96

  downloads

• 2

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.