Tweet
Hello Biomechanists,
Recently I've been reading some studies of the compliance of the
human aorta - studies that were intrinsically
physical in character, but that were conducted by medically qualified
people. The poor science that has been conducted deserves
commentary, but I will not name names. This is more in the way of a
cautionary note to those reading in the field!
Most of the studies are rendered almost useless by poor
experimental design and interpretation of the results - aortic
distensibility depends on age, the instantaneous degree to which
the aorta has already been stretched when its distensibility is
measured, the exact location of the measurement in the aorta
(compliance decreases tenfold from the aortic sinuses to the aortic
bifurcation), and the pressure fluctuation that has caused it to
distend - the pressure pulse changes its wave shape and amplitude
significantly from the aortic valve to the bifurcation. Typical
errors in the studies that have been published are:
1 To neglect the increase in mean blood pressure in older
patients, and thus the fact that the aorta is stretched tighter
before any change in distensibility due to age-related tissue
changes are taken into account. Not all apparent stiffening is
due to age.
2 To take measurements of distention all the way along the aorta, but
to subsequently calculate aortic distensibility from the blood pressure as
measured in the proximal ascending aorta, rather than locally.
3 To think that distensibility (compliance) is a constant value,
rather than one that varies radically with the hoop strain near
the limit of normal physiologic hoop strain.
4 To recognize changes in distensibility with hoop strain, but to
calculate the instantaneous hoop modulus on the basis of the
end-diastolic hoop stress, or hoop stress taking into account the
change in diameter, but ignoring the change in aortic wall
thickness.
One paper published by a group of German cardiologists and a single
engineer has compliance values that are incorrect by a factor of
roughly 16900! The fact that these values are repeated in the
discussion means that the authors have no idea of the magnitude of
stiffness and elasticity values. More worrying is the conclusion
that the referees didn't either. There is much misleading published
work - and there are some erroneous conclusions. In the
case of the aorta I have concluded that age-related changes in aortic
compliance are in fact mildly to greatly exaggerated by many
investigators, due to plain bad science. When stats is applied to such
bad data, of course the answer is highly statistically significant
nonsense. Perhaps this field needs to be "tightened up"?
Mark W Swanepoel, PhD
School of Mechanical Engineering
University of the Witwatersrand
South Africa
Tel: 0927 (0)11 716 2578
Fax: 0927 (0)11 339 7997
---------------------------------------------------------------
To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
For information and archives: http://isb.ri.ccf.org/biomch-l
---------------------------------------------------------------
Recently I've been reading some studies of the compliance of the
human aorta - studies that were intrinsically
physical in character, but that were conducted by medically qualified
people. The poor science that has been conducted deserves
commentary, but I will not name names. This is more in the way of a
cautionary note to those reading in the field!
Most of the studies are rendered almost useless by poor
experimental design and interpretation of the results - aortic
distensibility depends on age, the instantaneous degree to which
the aorta has already been stretched when its distensibility is
measured, the exact location of the measurement in the aorta
(compliance decreases tenfold from the aortic sinuses to the aortic
bifurcation), and the pressure fluctuation that has caused it to
distend - the pressure pulse changes its wave shape and amplitude
significantly from the aortic valve to the bifurcation. Typical
errors in the studies that have been published are:
1 To neglect the increase in mean blood pressure in older
patients, and thus the fact that the aorta is stretched tighter
before any change in distensibility due to age-related tissue
changes are taken into account. Not all apparent stiffening is
due to age.
2 To take measurements of distention all the way along the aorta, but
to subsequently calculate aortic distensibility from the blood pressure as
measured in the proximal ascending aorta, rather than locally.
3 To think that distensibility (compliance) is a constant value,
rather than one that varies radically with the hoop strain near
the limit of normal physiologic hoop strain.
4 To recognize changes in distensibility with hoop strain, but to
calculate the instantaneous hoop modulus on the basis of the
end-diastolic hoop stress, or hoop stress taking into account the
change in diameter, but ignoring the change in aortic wall
thickness.
One paper published by a group of German cardiologists and a single
engineer has compliance values that are incorrect by a factor of
roughly 16900! The fact that these values are repeated in the
discussion means that the authors have no idea of the magnitude of
stiffness and elasticity values. More worrying is the conclusion
that the referees didn't either. There is much misleading published
work - and there are some erroneous conclusions. In the
case of the aorta I have concluded that age-related changes in aortic
compliance are in fact mildly to greatly exaggerated by many
investigators, due to plain bad science. When stats is applied to such
bad data, of course the answer is highly statistically significant
nonsense. Perhaps this field needs to be "tightened up"?
Mark W Swanepoel, PhD
School of Mechanical Engineering
University of the Witwatersrand
South Africa
Tel: 0927 (0)11 716 2578
Fax: 0927 (0)11 339 7997
---------------------------------------------------------------
To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
For information and archives: http://isb.ri.ccf.org/biomch-l
---------------------------------------------------------------