Definition of ambulatory blood pressure targets for diagnosis and treatment of hypertension in relation to clinic blood pressure: prospective cohort study
BMJ 2010; 340 doi: https://doi.org/10.1136/bmj.c1104 (Published 14 April 2010) Cite this as: BMJ 2010;340:c1104
All rapid responses
The Australian study by Head and colleagues1 reiterates aspects of
ambulatory blood pressure measurement that were documented over 15 years
ago,2 namely: (1) office blood pressure measurements taken by doctors are
higher than those obtained by trained staff; (2) the white-coat effect
increases with higher levels of clinic blood pressure; (3) for
classification of different grades of hypertension thresholds based on the
ambulatory blood pressure are lower than those for office blood pressure.
The Australian report fails to acknowledge long-standing
recommendations on device validation,3 by using the Suntech Accutracker as
one of the devices for measuring ambulatory blood pressure; this device
received a C grade for diastolic blood pressure with the British
Hypertension Society protocol and is not recommended for clinical
practice.4,5 The Australian study included only referred patients and
self-selected normotensive subjects recruited by advertisement and has
therefore limited application in practice.
Thresholds for ambulatory blood pressure measurement were originally
derived from the upper limits of the distribution (means +/- 2 SDs or 95th
percentile) in subjects with office normotension.2 The regression
approach in the Australian study is a statistical method first applied in
the PAMELA study.6 However, using the regression approach ambulatory
thresholds were initially set at too low levels, because for the
extrapolation of the ambulatory blood pressure at given levels of the
office blood pressure the smaller confidence intervals for prediction of
the mean ambulatory blood pressure of the population were not
differentiated from the wider confidence intervals for the prediction of
an individual’s ambulatory blood pressure.7 The Australian study1 avoided
this problem by reporting only point estimates without confidence
intervals, which greatly weakens any statistical conclusions.
It took over 20 years to collect the necessary prospective data to
define outcome-driven thresholds.7 The table summarises how evidence
influenced the selection of ambulatory thresholds, beginning with those
proposed in the BMJ in 2001,9 then moving forward to ambulatory thresholds
based on statistical approaches (European Society of Hypertension [ESH]
200310 and 200711 guidelines) and finally arriving at lower thresholds
based on the 10-year cardiovascular risk from the International Database
on the Ambulatory blood pressure and Cardiovascular Outcomes [IDACO].8
In conclusion, the Australian report,1 which uses an outdated
statistical approach and ignores most of the literature published on the
subject, concludes by recommending an ambulatory daytime mean threshold of
136/87 mm Hg, but in so doing has failed to recognise that outcome-
thresholds are the gold standard for diagnostic criteria.
References
(1) Head GA, Milhailidou AS, Duggan KA, Beilin LJ, Berry N, Brown MA
et al. Definition of ambulatory blood pressure targets for diagnosis and
treatment of hypertension in relation to clinic blood pressure:
prospective cohort study. Br Med J. In press 2010.
(2) Staessen JA, O'Brien ET, Amery AK, Atkins N, Baumgart P, De Cort
P et al. Ambulatory blood pressure in normotensive and hypertensive
subjects : results from an international database. J Hypertens 1994; 12
(suppl 7):S1-S12.
(3) O'Brien E, Petrie J, Littler W, de Swiet M, Padfield PL, O'Malley
K et al. The British Hypertension Society protocol for the evaluation of
automated and semi-automated blood pressure measuring devices with special
reference to ambulatory systems. J Hypertens 1990; 8:607-619.
(4) Taylor R, Chidley K, Goodwin J, Broeders M, Kirby B. Accutracker
II (version 30/23) ambulatory blood pressure monitor : clinical validation
using the British Hypertensin Society and Association for the Advancement
of Medical Instrumentation standards. J Hypertens 1993; 11:1275-1282.
(5) See URL: htpp://www.dableducational.org (Accessed 21.04.2010)
(6) Mancia G, Sega R, Bravi C, Di Vito G, Valagussa F, Cesana G et
al. Ambulatory blood pressure normality : results from the PAMELA study. J
Hypertens 1995; 13:1377-1390.
(7) Staessen JA, Bieniaszewski L, O'Brien ET, Imai Y, Fagard R. An
epidemiological approach to ambulatory blood pressure monitoring : the
Belgian population study. Blood Press Monit 1996; 1:13-26.
(8) Kikuya M, Hansen TW, Thijs L, Björklund-Bodegård K, Kuznetsova T,
Ohkubo T et al. Diagnostic thresholds for ambulatory blood pressure
monitoring based on 10-year cardiovascular risk. Circulation 2007;
115:2145-2152.
(9) O'Brien E, Beevers G, Lip GY. ABC of hypertension. Blood pressure
measurement. Part III-automated sphygmomanometry: ambulatory blood
pressure measurement. Br Med J 2001; 322:1110-1114.
(10) European Society of Hypertension/European Society of Cardiology
Guidelines Committee. 2003 European Society of Hypertension/European
Society of Cardiology guidelines for the management of arterial
hypertension. J Hypertens 2003; 21:1011-1053.
(11) The Task Force for the Management of Arterial Hypertension of
the European Society of Hypertension (ESH) and of the European Society of
Cardiology (ESC), Mancia G, De Backer G, Dominiczak A, Cifkova R, Fagard R
et al. 2007 guidelines for the management of arterial hypertension. Eur
Heart J 2007; 28:1462-1536.
TABLE: Thresholds for ambulatory blood pressure measurement
Thresholds for ambulatory blood pressure measurement Source Optimal Normal Hypertension BMJ 2001[9] 24 hours ... ≤130/80 >135/85 daytime ... ≤135/85 >140/90 nighttime ... ≤120/70 >125/75 ESH 2003/2007[10,11] awake 140/90 asleep 125/65 IDACO 2008[8] 24 hours
Competing interests:
None declared
Competing interests: Thresholds for ambulatory blood pressure measurement Source Optimal Normal Hypertension BMJ 2001[9] 24 hours ... ≤130/80 >135/85daytime ... ≤135/85 >140/90nighttime ... ≤120/70 >125/75ESH 2003/2007[10,11] awake 140/90asleep 125/65IDACO 2008[8] 24 hours
Reply to: “Clinical thresholds for ambulatory blood pressure measurement reinvented?”
In their letter
in response to our recent article “Definition of ambulatory blood pressure
targets for diagnosis and treatment of hypertension in relation to clinic blood
pressure: a prospective cohort study” 1 Staessen and O’Brien raise a number of issues that
need to be addressed. Their call for target ambulatory blood pressures to be
based on outcome studies is one we would fully support, but ours was clearly
not an outcome study and nor was it designed to determine what should be the
targets for ambulatory blood pressures. Importantly, and to reflect usual practice,
we examined ambulatory blood pressure equivalents to clinic blood pressure
targets, measured by either doctors or trained healthcare professionals. The response by Staessen and O’Brien has
therefore missed the major novel findings of our trial which were the provision
of a complete set of ambulatory blood pressure equivalents: for the three
levels of hypertension, for well-defined target clinic blood pressures in
subjects with a single or multiple risk factors and also equivalents for
subjects with proteinuria. We further stratified these for gender and age as
well as providing analysis for seated and reclining clinic blood pressure.
The suggestion that the Suntec Accutracker is not
recommended for clinical practice is based on an early version of this device
as evaluated by O’Brien et al. 2, but this device did pass the
Association for the Advancement of Medical Instrumentation (AAMI) tests.2 Nevertheless, we have
reanalysed our data using only those devices that have passed both British and
USA guidelines. This involved removing only a small amount of data which made
no difference at all (see table 1).
Our study conclusions are not at all dependent on how the
subjects might be recruited or whether they were on or off treatment. We have not preselected or excluded
patients but used those referred for monitoring from a wide cross-section of
Australian general practices and hospital clinics and this has direct relevance
to clinical practice. There was a much bigger effect on the target blood
pressure depending on who took the clinic blood pressure and whether the
subjects were male or female/ old or young.
Finally, our regression approach is not the same as PAMELA 3 and is not outdated as we
used a least product regression technique which is novel in this field and is
the correct statistical approach 4. When both X and Y varies,
the standard least squares approach is not valid and the ordinary least product
regression should be used. The least squares line is far too shallow in light
of being biased by very high clinic blood pressures. Staessen and O’Brien refer
to studies that give confidence intervals but these will need to be reanalysed
as the intervals are around the wrong regression lines.
References
1.
Head G, Mihailidou A, Duggan K, Beilin L, Berry N, Brown M, et al. Relationship
between ambulatory and clinic blood pressure: Defining diagnostic and treatment
targets. Brit Med J 2010:in press.
2. O'Brien E,
Coats A, Owens P, Petrie J, Padfield PL, Littler WA, et al. Use and
interpretation of ambulatory blood pressure monitoring: recommendations of the
British hypertension society. Bmj
2000;320(7242):1128-1134.
3. Mancia G,
Sega R, Bravi C, De Vito G, Valagussa F, Cesana G, et al. Ambulatory blood
pressure normality: results from the PAMELA study. J Hypertens 1995;13(12 Pt 1):1377-90.
4. Ludbrook
J. Comparing methods of measurement. Clin
Exp Pharmacol Physiol 1997;24(2):193-203.
Table 1 Systolic/diastolic
ambulatory blood pressure (ABP) values predicted from seated clinic blood
pressure levels; values in mm Hg
Clinic blood
pressure threshold
ABP predicted
from staff measured seated clinic blood pressure all devices (n=5327)
ABP predicted
from staff measured seated clinic blood
pressure excluding Suntec Accutracker (n=4875)
24 hour
Night
Day
24 hour
Night
Day
Grade 3 (severe) hypertension
>180/110
163/101
157/93
168/105
162/101
156/93
166/105
Grade 2 (moderate) hypertension
>160/100
148/93
139/84
152/96
147/93
138/84
151/96
Grade 1 (mild) hypertension
>140/90
133/84
121/76
136/87
132/84
120/75
135/87
Target blood pressure plus one
condition
125/76
112/67
128/78
124/75
111/67
127/78
Target blood pressure with
proteinuria
121/71
107/63
124/74
120/71
106/62
123/73
Normal blood pressure
117/76
102/67
120/78
117/75
102/67
120/78
Competing interests:
None declared
All of the authors of the original article are also authors of this response.
Competing interests: The suggestion that the Suntec Accutracker is notrecommended for clinical practice is based on an early version of this deviceas evaluated by O’Brien et al.