Ambulatory Tonometric Blood Pressure Measurements
in Patients with Diabetes
Simone Theilade, M.D., Christel Joergensen, M.D., Frederik Persson, M.D., Maria Lajer, Ph.D.,
and Peter Rossing, DmSc
Background: Arterial tonometry is a novel technique for measuring ambulatory blood pressure (AMBP). The watch-like
(HealthSTATS International, Singapore) captures radial pulsewave reﬂection and calculates brachial blood
pressure (BP). In this study we investigate if arterial tonometry is applicable and reliable in patients with diabetes.
Subjects and Methods: We compared tonometric (BPro) to cuff-based oscillometric and auscultatoric BPs (Takeda model
TM2421, A&D Medical, Tokyo, Japan) in 25 Caucasian patients with type 1 or type 2 diabetes. Patients were seen twice within
2 weeks. At visit 1, a 15-min rest was followed by the recording of three cuff-based BPs and 2-min continuous tonometric BPs.
At both visits 24-h AMBP measurements were recorded with the BPro device.
Results: At Visit 1, auscultatoric BP (mean –SD) was 136 –19/72 –8 mm Hg versus 138 –19/78 –8 mm Hg with the tono-
metric device. Visit 1 AMBP was 131 –20/76 –9 mm Hg versus 131 –12/75 –9 mm Hg at Visit 2. Mean 24-h AMBP, daytime
BP, nighttime BP, and dipping at the two visits were similar (P>0.40). Linear and intraclass correlations coefﬁcients between
auscultatoric and tonometric systolic and diastolic BP were r=0.86 and 0.65, respectively (P<0.001 for both), and r=0.83 and
0.77, respectively (P<0.001 for both). The mean differences between devices were 1.9 –10 and 5.5 –6.6 mm Hg for systolic and
diastolic BP, respectively.
Conclusions: In patients with diabetes tonometric and cuff-based BPs are comparable, and tonometric AMBPs are repro-
ducible and feasible.
Blood pressure (BP) control is paramount in preven-
tion of complications to diabetes, and hypertension
signiﬁcantly increases macro- and microvascular compli-
ABPof ‡130 mm Hg deﬁnes hypertension in
Conventionally, BP has been measured with sphygmo-
manometry, using a device consisting of an inﬂatable cuff for
occlusion of blood ﬂow and a manometer for measuring
In recent years arterial tonometry has emerged as
an alternative way of indirect BP measuring. A tonometric
device captures pulsewaves from arterial surfaces (e.g.,
the radial artery) and translate wave characteristics to BPs.
Oscillometric devices are less accurate in measuring BP in
persons with increased arterial stiffness.
It is possible that
increased arterial stiffness also impair the accuracy of tono-
metric BP measurements. Patients with diabetes have in-
creased arterial stiffness,
and with this study we aimed to
investigate if tonometry is applicable and reliable in BP
measuring in patients with diabetes.
Some of these data have been previously published in ab-
Subjects and Methods
The study design is cross-sectional, including 25 patients: 17
with type 1 and eight with type 2 diabetes. All were Caucasian.
Patients were 59.6 –7.8 years old, with a duration of diabetes
of 24.5 –14.3 years, and 18 (72%) men. Nine patients had
normoalbuminuria, deﬁned as persistent normoalbuminuria
with a urinary albumin excretion rate of <30 mg/24-h period.
Two and 14 patients had micro- and macroalbuminuria,
respectively, deﬁned as a urinary albumin excretion rate be-
tween 30 and 300 mg/24-h period and >300 mg/24-h period
in two out of three consecutive measurements. Glomerular
Steno Diabetes Center, Gentofte, Denmark.
Data were previously presented in abstract form at the 2011 meeting of the European Association for the Study of Diabetes and the 2011
meeting of the Danish Society of Hypertension.
DIABETES TECHNOLOGY & THERAPEUTICS
Volume 14, Number 6, 2012
ªMary Ann Liebert, Inc.
ﬁltration rate (GFR) in patients with macroalbuminuria was
73.6 –34.2 mL/min/1.73 m
All patients were followed up at Steno Diabetes Center,
Gentofte, Denmark. They were randomly selected from the
outpatient clinic without regard for sex, age, type of diabetes,
or kidney function. The number of participants included in
the study was based on power calculations. One patient was
excluded from the validation analysis because of missing
sphygmomanometric measurements. Another patient only
had one valid 24-h ambulatory BP (AMBP) measurement and
was not included in the reproducibility analysis.
Sphygmomanometric BPs were recorded with a Takeda
device (TM2421, version 7, A&D Medical, Tokyo, Japan) and
tonometric measurements with the BPro
International, Singapore) device (Fig. 1). Both Takeda and
BPro measurements were recorded on the left arm on all pa-
tients. The Takeda device is currently used for AMBP mea-
surements at Steno Diabetes Center and was therefore chosen
for comparison with the BPro device. Measurements of both
cuff-based and tonometric BPs were performed by trained
The cuff-based Takeda device obtains auscultatoric and
oscillometric BP measurements simultaneously. Auscultatoric
systolic BP (SBP) and diastolic BP (DBP) measurements are
recorded at the appearance of the ﬁrst and by the disap-
pearance of the last Korotkoff sound, when entering the ﬁfth
Korotkoff’s phase, respectively. The transducer records os-
cillations equivalent to mean arterial pressure (MAP) and
calculates oscillometric BPs based on accepted algorithms.
Auscultatoric SBPs are typically lower, whereas DBPs are
higher, than oscillometric measurements.
The BPro is a watch-like device that captures radial pulse-
wave reﬂection and calculates brachial AMBP. A brachial
AMBP can be recorded for every 10 s of pulsewave mea-
surements. The BPro has previously been validated and meets
European Society of Hypertension and Association for the
Advancement of Medical Instrumentation standards.
study, the BPro device was calibrated with an oscillometric
device (model UA 787, A&D Medical) prior to BP measuring,
and every other year the device is serviced by the manufac-
At Visit 1, patients were placed supine in a resting position
for 15 min. A cuff of appropriate size was placed on the right
arm. The stethoscope for auscultatoric and the transducer
for oscillometric measurements were both placed above the
brachial artery. Three separate Takeda measurements were
obtained, and subsequently 2-min continuous BPro mea-
surements were recorded.
At both visits AMBP was recorded with BPro. AMBP re-
cordings were performed every 15 min during daytime and
nighttime. Daytime AMBP was from 7 a.m. to 11 p.m.,
whereas nighttime AMBP was 11 p.m. to 7 a.m. The BPro
device also calculated dipping, which refers to the percentage
decline in BP from daytime to nighttime. Mean AMBP was
calculated as the average of all measurements obtained
during the 24-h period.
Statistical analysis is performed using SPSS version 15.0 for
Windows (SPSS Inc., Chicago, IL). Variables are given as
mean –SD values. Correlations between devices are calcu-
lated by linear regression equations and intraclass correlation.
Mean differences between devices are calculated and visual-
ized in a Bland–Altman plot, and 95% limits of agreement
are calculated. Paired Student’s ttests compare AMBPs, and
independent-samples ttest compares groups. A two-tailed
Pvalue of <0.05 is considered statistically signiﬁcant.
The study conformed to the Declaration of Helsinki, and all
patients gave informed consent.
Auscultatoric, oscillometric, and tonometric BPs (mean –
SD) were 136 –19/72 –8, 137 –16/77 –9, and 138 –19/78 –
8 mm Hg, respectively.
Respective linear correlation coefﬁcients between auscul-
tatoric and tonometric SBP and DBP were r=0.86 and 0.65
(P<0.001 for both) (Fig. 2a), with intraclass correlations co-
efﬁcients of r=0.83 and 0.77 (P<0.001 for both). Respective
linear correlation coefﬁcients between oscillometric and to-
nometric SBP and DBP were r=0.89 and 0.82 (P<0.001 for
both), with intraclass correlation coefﬁcients of r=0.94 and
0.90 (P<0.001 for both). Respective linear correlation coefﬁ-
cients between auscultatoric and oscillometric SBP and DBP
were r=0.91 and 0.77 (P<0.001 for both), with intraclass
correlation coefﬁcients of r=0.86 and 0.87 (P<0.001 for both).
Mean differences between tonometric and auscultatoric
SBP and DBP were 1.9 –10 and 5.5 –6.6 mm Hg, respectively,
with 95% limits of agreement of -17.7 to 21.5 and -7.4 to
18.4 mm Hg, respectively (Fig. 2b). Mean differences between
tonometric and oscillometric SBP and DBP were 1.9 –8.6 and
0.0 –5.1 mm Hg, respectively, with 95% limits of agreement of
-15.0 to 18.8 and -10 to 10 mm Hg, respectively. Mean dif-
ferences between auscultatoric and oscillometric SBP and
DBP were 0.2 –8.0 and 5.6 –5.8 mm Hg, respectively, with
95% limits of agreement of -15.5 to 15.9 and -5.8 to 17.0 mm
FIG. 1. Photograph of the BPro.
454 THEILADE ET AL.
There were no signiﬁcant differences in agreement between
devices when comparing normo- versus micro- and macro-
albuminuric patients, men versus women, or patients with
type 1 versus type 2 diabetes (P>0.05) (data not shown).
Evaluation of repeated measurements of AMBP
Mean BP values from the two visits were similar. Mean
AMBP was 131 –20/76 –9 mm Hg versus 132 –12/75 –9mm
Hg (P=0.917 and P=0.954, respectively), mean daytime BP
was 136 –22/78 –9 mm Hg versus 136 –13/78 –10 mm Hg
(P=0.385 and P=0.972, respectively), mean nighttime BP was
124 –20/71 –9 mm Hg versus 124 –13/71 –8mm Hg (P=
0.935 and P=0.984, respectively), and mean dipping was
9.3 –5.7% versus 9.1 –4.8% (P=0.871).
Patient satisfaction and AMBP completion
Only one patient (4%) declined to repeat the AMBP. For the
remaining patients the discomfort associated with the tono-
metric AMBP was acceptable, and AMBP measurements were
successful according to present guidelines.
In this study on patients with diabetes, we found tono-
metric and cuff-based BPs to be comparable and tonometric
AMBPs to be reproducible. The BPro values correlated better
with oscillometric than with auscultatoric values, which could
be explained by the BPro device being calibrated with an
We obtained lower SBP and DBP auscultatoric than oscil-
lometric values, despite oscillometric values typically over-
estimating SBP and underestimating DBP. This could be
caused by our cohort having increased arterial stiffness sec-
ondary to diabetes,
leading to overestimation of both SBP
and DBP oscillometric values and underestimation of aus-
We did not ﬁnd any signiﬁcant difference in agreement
between devices when subdividing patients according to al-
buminuria status, sex or type of diabetes. However, this
ﬁnding may be due to lack of power.
Aside from brachial arterial BP, the BPro device measures
various indices of arterial stiffness including augmentation
index, central BP, and MAP. Brachial arterial BP is an inferior
risk marker of cardiovascular outcome compared with mark-
ers of arterial stiffness.
As patients with diabetes have
increased arterial stiffness,
measuring indices of arterial
stiffness rather than peripheral BP may prove to be a better risk
marker as well as a better treatment target for management of
hypertension in patients with diabetes. Therefore the BPro
possess potentially advantageous qualities for the purpose of
measurements of both BP and arterial stiffness, although we
did not investigate arterial stiffness in the current study.
A great advantage of the BPro is the smaller size and more
accessible position of the device, causing less discomfort both
during and between BP measurements. The absence of an
inﬂatable cuff and the inaudibility of the device prevent an-
ticipation rise of BP and nighttime awakening in connection
with measurements. This allows for more frequent and un-
disturbed measuring and possibly a more accurate AMBP.
Our study has some limitations. The small number of
participants does not allow for scrutinizing patient data when
dividing patients according to kidney function, gender, or
type of diabetes. However, we did not see any association
between patient characteristics and reliability of tonometric
BPs. We only performed tonometric AMBPs. Concurrent
AMBP measurements with both cuff-based and tonometric
devices would have been interesting. However, as both de-
vices would have had to be strapped on to the same arm,
faulty measurements were likely to be obtained. Furthermore,
the application of two devices concurrently probably
would have been bothersome to patients, thereby affecting
FIG. 2. (a) Correlations plot and (b) Bland–Altman plot for systolic blood pressure (SBP) measured with sphygmoma-
nometry (auscultatoric) (Takeda) versus tonometry (BPro).
TONOMETRIC BP MEASUREMENTS IN DIABETES 455
In conclusion, we ﬁnd that the manufacturer-validated
BPro device recorded reliable and reproducible tonometric BP
measurements in patients with diabetes.
The BPro device offers frequent and undisturbed BP mea-
surements, along with additional information on arterial
state. Furthermore, tonometric measurements appear to be
feasible and possibly more convenient for the patients. We
therefore propose that in patients with diabetes tonometric
BPs may be preferable to sphygmomanometric measure-
However, larger studies with sufﬁcient follow-up are
needed to establish the role of tonometric BP measurements in
diabetes, in particular for the purpose of evaluating the value
of parameters of arterial stiffness.
Author Disclosure Statement
No competing ﬁnancial interests exist.
1. Adler AI, Stratton IM, Neil HA, Yudkin JS, Matthews DR,
Cull CA, Wright AD, Turner RC, Holman RR: Association of
systolic blood pressure with macrovascular and microvas-
cular complications of type 2 diabetes (UKPDS 36): pro-
spective observational study. BMJ 2000;321:412–419.
2. Chen G, McAlister FA, Walker RL, Hemmelgarn BR,
Campbell NR: Cardiovascular outcomes in Framingham
participants with diabetes: the importance of blood pressure.
3. American Diabetes Association: Standards of medical care in
diabetes—2010. Diabetes Care 2010;33(Suppl 1):S11–S61.
4. Pickering TG, Hall JE, Appel LJ, Falkner BE, Graves J, Hill
MN, Jones DW, Kurtz T, Sheps SG, Roccella EJ: Rec-
ommendations for blood pressure measurement in humans
and experimental animals: Part 1: Blood pressure measure-
ment in humans: a statement for professionals from the
Subcommittee of Professional and Public Education of the
American Heart Association Council on High Blood Pres-
sure Research. Circulation 2005;111:697–716.
5. Chen CH, Nevo E, Fetics B, Pak PH, Yin FC, Maughan WL,
Kass DA: Estimation of central aortic pressure waveform by
mathematical transformation of radial tonometry pressure.
Validation of generalized transfer function. Circulation
6. van Popele NM, Bos WJ, de Beer NA, van Der Kuip DA,
Hofman A, Grobbee DE, Witteman JC: Arterial stiffness as
underlying mechanism of disagreement between an oscil-
lometric blood pressure monitor and a sphygmomanometer.
7. Gordin D, Waden J, Forsblom C, Thorn LM, Rosengard-
Barlund M, Heikkila O, Saraheimo M, Tolonen N, Hietala K,
Soro-Paavonen A, Salovaara L, Makinen VP, Peltola T,
Bernardi L, Groop PH: Arterial stiffness and vascular com-
plications in patients with type 1 diabetes: the Finnish Dia-
betic Nephropathy (FinnDiane) Study. Ann Med 2012;44:
8. Theilade S, Lajer M, Joergensen C, Persson F, Rossing P:
Tonometric ambulatory blood pressure measurements are
reliable and improve hypertension diagnostics in patients
with type 1 diabetes [abstract]. Diabetologia 2011;54(Suppl 1):
9. Nair D, Tan SY, Gan HW, Lim SF, Tan J, Zhu M, Gao H,
Chua NH, Peh WL, Mak KH: The use of ambulatory tono-
metric radial arterial wave capture to measure ambulatory
blood pressure: the validation of a novel wrist-bound device
in adults. J Hum Hypertens 2008;22:220–222.
10. O’Brien E, Asmar R, Beilin L, Imai Y, Mallion JM, Mancia G,
Mengden T, Myers M, Padﬁeld P, Palatini P, Parati G,
Pickering T, Redon J, Staessen J, Stergiou G, Verdecchia P:
European Society of Hypertension recommendations for
conventional, ambulatory and home blood pressure mea-
surement. J Hypertens 2003;21:821–848.
11. Boutouyrie P, Tropeano AI, Asmar R, Gautier I, Benetos A,
Lacolley P, Laurant S: Aortic stiffness is an independent
predictor of primary coronary events in hypertensive pa-
tients: a longitudinal study. Hypertension 2002;39:10–15.
Address correspondence to:
Simone Theilade, M.D.
Steno Diabetes Center
Niels Steensens Vej 2, NLC 2.07
DK-2820 Gentofte, Denmark
456 THEILADE ET AL.