|
Professor Michell graduated as a
veterinary surgeon but, following
postdoctoral experience in various American medical institutes,
has pursued a career in caparative medicine in which he holds a
personal chair at the University of London (St Bartholomew’s
Hospital)
e-mail b.michell@qmul.ac.uk
|
A single reading is practically useless |
Doctors are not good at measuring blood pressure, even though it is
important. Nurses are generally better; they have more patience — they
repeat readings whereas doctors regard themselves as too busy. Yet single
readings
are practically useless and hypertension, with its consequences, is a
huge drain on NHS budgets. The procedure, sanctified by antiquity, is
more like a religious ceremony than a scientifically validated clinical
technique. This is bizarre considering the ease and sophistication with
which BP can be measured, even throughout 24 hours, given the will and
the equipment to do it properly. BP measurement is set to change — hopefully
to catch up with modern standards of reproducibility (precision) — as
tighter laws on toxic hazards may lead to the elimination of the venerable
mercury sphygmomanometer, second only to the stethoscope in the doctor’s
traditional regalia.
Blood pressure is the oldest surviving instrumented medical measurement,
predating the clinical thermometer, so it is time that it was done properly.
The emergence of pharmacists as front-line troops in the battle to spot
and control hypertension early, before the damage begins silently and
asymptomatically, is appropriate since they are well placed to achieve
progress in three key areas: accurate measurement, astute evaluation
of side effects and maximisation of compliance. All of these are currently
achieved only haphazardly, resulting in great personal cost to patients
and financial cost to the health budget.
Inadequate measurements cost lives
There used to be a war-time poster campaign “careless talk costs
lives”; there should be a similar campaign concerning careless
blood pressure measurement. It remains true that only a minority of people
with hypertension are correctly diagnosed and of these, only a minority
are properly treated. Most, for example, require more than one medicine
to regain the normal range.
Inadequate measurement creates three unnecessary sources of hypertensive
damage or cost: premature use of medication among those who are not hypertensive,
lack of medication among those who are, and flawed evaluation of response
to therapy, leading to misguided decisions on changes of medication.
It is hard enough to control blood pressure over decades of patients’ lives,
often in the early stages with drugs that cause more discomfort than
the underlying condition. It is even harder if the beacon measurement
routinely remains a haphazard shot in the dark. There is a clear opportunity
for pharmacists not just to relieve doctors of unnecessary work, but
to set an example by doing properly the measurement which is so poorly
performed in most GP surgeries.1,2 What do we need to measure?
If we ask “what is normal blood pressure?” or “what
matters about abnormal blood pressure?” the answers remain surprisingly
mysterious. Of course, physiologically, we are taught arterial pressure
regulation as a great example of negative feedback ensuring homeostatic
constancy at the “set-point”, like body temperature. If arterial
pressure falls, cardiac output or peripheral resistance are increased
to restore it to “normal”, just like thermostats and central
heating.
If Reverend Stephen Hales time-warped from the 18th century into a physiology
class he would be bewildered by this idea. If his famous pioneering measurements
from horses, and especially from dogs, are converted from feet of blood
into modern measurements, they are remarkably authentic. “If it
is so perfectly regulated,” he might have asked “why is it
so remarkably variable?” Which it is, as he observed and as we
now know, not only hour by hour, not just because of distractions, but
even minute by minute. Perhaps the most mysterious aspect of blood pressure
is the fact that something so variable, so carelessly measured, remains
characteristic of individual patients; if a human cohort is followed
as it matures over the decades there is “tracking” of blood
pressure, ie, the rank order of the individuals remains similar; similar
evidence exists for dogs, although over much shorter timespans.
Since blood pressure is inherently variable, at least five readings should
be averaged, excluding outliers. The first reading probably serves to
seat the cuff so it should be discarded. It is also important that the
cuff is approximately level with the heart, is the correct size in relation
to the circumference of the arm and that the measuring device has been
properly validated, calibrated and maintained. Ideally, time of day should
be standardised for an individual, because of the diurnal rhythm of BP,
and the patient must be given time (ideally five minutes) to settle before
measurements begin; even slight anxiety elevates BP (the white coat effect).
Such precautions, routinely applied to blood pressure measurement in
dogs, yield much more reproducible readings in humans than those obtained
by the usual clinical routine.1–3 Sadly, the current recommendation
of the British Hypertension
Society is still to rely on merely two readings;4 you might as well define the depth of the Thames at Westminster with
two consecutive readings at unspecified times. The guidelines do, however,
acknowledge the unreliability of the first reading (still the only reading
taken by many GPs).
Recently, with the growth in home measurement and ambulatory blood pressure
measurement, the concept of “masked” hypertension has emerged.5 This is the opposite of white coat hypertension (where pressure is elevated
in the clinic). Masked hypertension is particularly damaging because
pressure appears to be normal in the clinic but is actually higher elsewhere,
hence by escaping detection masked hypertension makes patients more vulnerable
to consequential damage. A key predisposing factor in masked hypertension
is simply taking an insufficient number of measurements in the clinic.6 In each cardiac cycle, arterial pressure reaches a peak (systolic) and
a trough (diastolic); these, together with mean arterial pressure are
the three usual measurements. Mean is not simply an average of the other
two but reflects the change in pressure over time during the cycle. Since
systole is fast whereas diastole is slow, it lies nearer to diastolic
pressure. When heart rate increases, diastole shortens so mean arterial
pressure increases. With increased BP, there is controversy as to which
does more harm, systolic or diastolic pressure, and it may well be the
difference between them, the pulse pressure, which matters most.
Hypertension in veterinary medicine
With the Government’s wish to see pharmacists routinely dispensing
veterinary prescriptions, pharmacies will soon be dealing with hypertensive
cats and dogs, as well as humans. Essential hypertension, ie, primary
hypertension (no other apparent cause), predominates in humans but
in cats and dogs hypertension is usually secondary to underlying
disease, especially renal disease in cats, diabetes and Cushing’s
disease in dogs; it also occurs in canine chronic renal failure
but much less often than in humans. In cats the clinical presentation
is often blindness due to retinal haemorrhage.
For measurements in dogs, the best equipment, because it has been
properly validated in published peer-reviewed papers, is the “DINAMAP” oscillometric
monitor, using the correct sized cuff on the tail base.7 It is much
harder to use in cats, with the cuff needing to be applied ahead
of connection to the machine, and plenty of time to settle — even
then some cats will not give a reading. Hence, although oscillometric
techniques display the three routine pressures and heart rate,
many prefer the Doppler technique for cats. Even though this only
gives
systolic pressure, it provides clear indications of moderate or
severe hypertension. These indirect techniques avoid arterial puncture
which
is quite unsuitable for conscious animals, although direct measurement
is preferable for anaesthetised animals because it is more accurate,
ie, reads closer to the true (intra-aortic) value. In conscious
animals accuracy matters less than precision, so that departures
from the
normal range for that technique are dependably detected.
With dogs, any definition of hypertension has to take account of
the considerable breed differences in normal pressure; the normal
range for Labradors for example, is different from that for “sight
hounds”, eg, deerhounds, which have far higher pressures.
Why they have these higher pressures, especially the relevant genetic
determinants, and why they are apparently unharmed is an intriguing
focus of research; if they really are unharmed they are not truly
hypertensive. |
Blood
pressure rises with age but this probably results from excessive dietary
sodium intake. This should be a matter for serious concern since
the majority of salt intake is passive rather than discretionary. The
slope of the rise for any individual could well predestine whether he
or she will need antihypertensive medication during later decades. Blood,
of course, has many pressures — venous, capillary, arterial — and
even with arterial, it differs according to the site of measurement.
In peripheral arteries, compared with the aorta, it is augmented by interaction
with pressure waves reflected from branch points in the arterial tree.
Some tall, fit, athletic teenagers have abnormally high pressures, measured
in their arms, but their central (aortic) pressures are normal so they
are unharmed. In contrast, people with diabetes are at risk even with
normal arterial pressures. Diabetes is the biggest single cause of the
need for renal transplants: diabetic nephropathy is substantially a consequence
of hypertensive renal damage, even at normal blood pressures. Afferent
arterioles, which supply the glomeruli, are abnormally relaxed in diabetics.
Normal pressure in the glomeruli (although higher than other capillaries)
is well below arterial pressure, because pressure falls across the resistance
offered by the afferent arterioles. If they are too slack, that protection
is lost; glomeruli experience abnormally high pressures even when arterial
pressure remains normal. People with diabetes therefore need to keep
their blood pressure at the lower end of the normal range.
Since populations show a normal distribution of BP (although not quite
bell shaped), there are inevitably individuals at either extreme. It
does not necessarily harm them, any more than being seven feet tall.
Hence hypertension, meaning a potentially damaging elevation of BP, is
not defined statistically (eg, two standard deviations from average)
but by the threshold above which the risk of adverse effects increases,
eg, damage to retinal, coronary, renal and cerebral blood vessels (hence
strokes). Normal human BP is routinely regarded as 140/90 (systolic/diastolic)
but there is recent evidence that even below this, higher pressures are
less healthy.
We have much to learn and the bedrock of progress will be proper measurement,
with all the patience which our patients deserve. No other human disease
comes close to the combined costs of treating hypertension and the consequential
costs of failing to do so adequately.
References
1. Michell AR. Routine blood pressure measurement: application of the
standard canine technique in a human. Blood Pressure Monitoring 1996;1:385–7.
2. Michell AR. Blood pressure measurement: single measurements would
not withstand 21st century peer review. BMJ 2001;323:805–6
3. Michell AR. Blood pressure measurement and hypertension in dogs. In
Practice 2000;22:274–5.
4. Williams B, Poulter NR, Brown MJ, Davis M, McInnes GT, Potter JF et
al. Guidelines for management of hypertension: report of the fourth working
party of the British Hypertension Society, 2004-BHS IV. Journal of Human
Hypertension 2004;18:139–85.
5. Palatini P. Masked hypertension: how can the condition be detected?
Blood Pressure Measurement 2004;9:297–9.
6. Mallion JM, Genes N, Vaur L, Clerson P, Vaisse B, Bobrie G et al.
Detection of masked hypertension by home blood pressure measurement;
is the number of measurements an important issue? Blood Pressure Measurement
2004;9:301–6.
7. Bodey AR, Michell AR. Reproducibility and variability of indirect
(oscillometric) blood pressure measurements in dogs: longitudinal studies.
Research in Veterinary Science 1996;63:15–22. |
The
Pharmaceutical Journal