Journal of Human Hypertension (2000) 14, 111–115
 2000 Macmillan Publishers Ltd All rights reserved 0950-9240/00 $15.00
www.nature.com/jhh
ORIGINAL ARTICLE
Blood pressure control in subjects with
type 2 diabetes
JA Hänninen1, JK Takala2 and SM Keinänen-Kiukaanniemi3
1
Health Centre of Mikkeli, Kiiskinmäenk 5–7, Mikkeli; 2Department of Community Health and General
Practice, University of Kuopio; 3Department of Public Health Science and General Practice and Oulu
University Hospital, University of Oulu, Aapistie 1, FIN-90220 Oulu, Finland
Blood pressure (BP) control of type 2 diabetic subjects
aged under 65 years was assessed in a primary care
setting. In addition, the usefulness of 24-h ambulatory
BP measurement (ABPM) in the treatment of hypertension was assessed in subjects with diastolic BP (DBP)
⭓90 mm Hg. Of the total 381 diabetic subjects, 260 (68%)
participated in the first phase, and 48 of the 110 subjects
with DBP ⭓90 mm Hg were equipped with a Meditech
ABPM-02 monitor in the second phase. The mean BP
of the 260 participants was 156/91 (s.d. 22/11) mm Hg.
According to the WHO criteria, 58% had hypertension,
and 42% had a diagnosis of hypertension. Albuminuria
⭓20 ␮g/min was detected in 32% of the subjects. Ten
percent of the subjects with diagnosed hypertension
had a mean BP ⬍140/90 mm Hg and 50% had a mean
BP ⭓160/95 mm Hg, as many as 38% of those not having
a diagnosis of hypertension. Only long-term poor BP
control in casual measurements was associated with
albuminuria (42% vs 27%, P ⴝ 0.018). It is concluded
that BP control was unsatisfactory and diagnosis of
hypertension was delayed in most subjects with type 2
diabetes. Occurrence of microalbuminuria was associated with poor BP control and urinary albumin excretion
rate may be useful in assessing the BP control. Further
studies are needed to assess the position of 24-h ABPM
in the treatment of hypertension of subjects with type
2 diabetes.
Journal of Human Hypertension (2000) 14, 111–115
Keywords: NIDDM; blood pressure control; albuminuria; quality of care; ABPM
Introduction
Elevated blood pressure (BP) is known to be a major
risk factor for mortality and complications in subjects with type 2 diabetes. This has been demonstrated in recent large prospective HOT and UKPDS
studies.1,2 These studies revealed that adequate BP
control not only reduced markedly microvascular
complications, but also overall and cardiovascular
mortality. From 38 to 66% of newly diagnosed type
2 diabetic patients have hypertension.1,3 The UK
Prospective Diabetes Study emphasised that reducing BP has a high priority in type 2 diabetes care.1
It has been stated, that BP control is poor in hypertensive subjects.4 –6 In a recent Finnish study of
hypertension care, good BP control (⬍140/90
mm Hg) was achieved in 11% of hypertensive subjects and acceptable control (⬍160/95 mm Hg) in
40% of 3520 hypertensive subjects treated in primary health care.4 Other studies have shown similar
treatment outcomes.5,6 The BP control has not been
a major outcome in diabetes care audits.7,8
An elevated urinary albumin excretion rate
(UAER) has been associated with increased mortality in subjects with type 2 diabetes.9–11 On the
other hand, high BP has increased the UAER in pro-
Correspondence: Jouko A Hänninen, Health Centre of Mikkeli,
Kiiskinmäenk. 5–7, FIN-50130 Mikkeli, Finland
Received 1 June 1999; revised 20 August 1999; accepted 27
September 1999
spective studies.12–14 In 24-h ambulatory BP
measurement (ABPM), those with white coat hypertension have had a decreased UAER15,16 and more
of those with an increased UAER have had the nondipper phenomenon,17 which supports the hypothesis that BP control is associated with urinary albumin excretion. Of the hypertensive type 2 diabetic
subjects, 8–23% have white coat hypertension16 and
30–31% show the non-dipper phenomenon.18 The
24-h ABPM could be useful in the treatment of
hypertension of subjects with type 2 diabetes to
detect the ones with white coat hypertension (lowrisk) and the non-dipper phenomenon (high-risk).
The problem is that a great number of diabetic
subjects have unsatisfactory BP control in casual
measurements. As the elevated UAER is associated
with higher BP, it could be used as an indicator of
BP control. The aim of the present study was to
assess BP control, and to test the usefulness of 24-h
ABPM in the management of hypertension of subjects with type 2 diabetes.
Materials and methods
The subjects of this study were type 2 diabetic
patients under 65 years of age in the area served by
the health centre of Mikkeli, in eastern Finland.
There were a total of 53 000 inhabitants, of which
42 000 were under 65 years old. Subjects with type 2
diabetes were identified with repeated fasting blood
glucose ⭓6.7 mmol/l and/or type 2 diabetes dia-
Blood pressure control in type 2 diabetes
JA Hänninen et al
112
gnosis. The formation of the study population has
been decribed previously.8
The study was carried out from December 1992 to
January 1994. Authorization for the study was provided by the ethical committee of the University
Hospital of Kuopio. Data for age, gender, number of
checkups, initial hypertension, antihypertensive
agents and initial coronary heart disease were
checked on a questionnaire supplemented from
patient records. BP level was determined with up to
five previous (during the last 2 years) BP measurements. In the health centre, BP was measured from
the right arm with a standard mercury sphygmomanometer by a trained nurse after at least 10 min
rest. Hypertension was defined as mean BP ⭓160/95
and/or antihypertensive treatment according to
WHO criteria.19 Good, satisfactory and poor BP control was defined according to the guidelines.20
Up to three fasting blood glucose values (during
the previous 2 years) were obtained from patient records and the mean of these values was used as a
measure of blood glucose. In the clinical examination, body mass index (BMI) was measured and
fasting blood sample was obtained to measure glycated haemoglobin A1c (HbA1c), total cholesterol,
high density lipoprotein (HDL) cholesterol and triglycerides. An overnight (12 h) urine sample was
collected and albumin excretion rate was determined. A single value ⭓20 ␮g/min was regarded as
albuminuria. Laboratory methods have been
presented previously in detail.8
Those subjects with mean diastolic BP (DBP) ⭓90
mm Hg were asked to participate in the 24-h ABPM.
This was performed with the Meditech ABPM-02
24-h fully-automatic, portable ambulatory BP monitor, which has been validated in 1994. According to
the criteria of British Hypertension Society, it has
been graded C for both diastolic and systolic values
against standard zero and C for systolic (SBP) and B
for diastolic (DBP) against random zero sphygmomanometer.21 The device was set to record BP reading at 20-min intervals between 7.00 am and 11.00
pm and at 40-min intervals during the night. Recordings were always started at 7.00 am and the subjects carried on with their normal daily activities
during the 24-h measurement. The data were analysed to calculate mean office time (8.00 am to 4.00
pm), daytime (7.00 am to 11.00 pm) and night-time
(11.00 pm to 7.00 am) BP values and daytime and
night-time mean arterial pressures (MAP).
White coat hypertension was defined as mean
office time BP ⬍85 mm Hg in 24-h ABPM and DBP
⭓90 mm Hg in casual measurements. The non-dipper phenomenon was defined as ⬍5 mm Hg mean
DBP decrease in night-time DBP compared to daytime DBP.
The study population was followed for 5 years
(September 1992 to September 1997) and those who
had died during the 5-year period were identified in
September 1997. Those who had moved during the
follow-up period were excluded from further analyses.
Statistical analyses were performed with SAS and
BMDP statistical software in the University of Oulu.
For nominal and ordinal variables, chi-square test
Journal of Human Hypertension
was used. For interval and ratio variables, two-tailed
Student’s t-test and Wilcoxon test were used when
appropriate. Survival rates were examined with Cox
regression analysis. To analyse how great differences in the UAER are detectable based on the
ABPM data with 90% degree of power, a logarithmic
transformation of UAER variable was performed (to
make it normally distributed) and the difference was
calculated from the formula
␦ = [(␴) *(z1−␣/2 + z1−␤)]/√n ,
where ␦ = differences in urinary albumin excretion
rate, ␴ = standard deviation, n = sample size, ␣ = Pvalue and ␤ = degree of power, and Z1−␣/2 is a 1−␣/2
fractile of normal distribution curve. The difference
in the urinary albumin excretion rate was estimated
by the 101±␦ (anti Ig) in the ABPM groups with the
approximation that the median urinary albumin
excretion rate was 101 = 10 ␮g/min.
Results
Of the 381 subjects aged under 65 years identified
with type 2 diabetes, 260 (68%) took part in the
study. The loss of the subjects has been described
in detail previously.16
In the total study population, the mean age was
56.3 years (s.d. 6.9), median duration of diabetes 6
years (range ⬍1–33), mean BMI 30.4 (s.d. 5.5) kg/m2
and mean HbA1c 8.6 (s.d. 2.0) %. The mean BP was
156/91 (s.d. 22/11) mm Hg, and 42% of the subjects
(108/260) had a diagnosis of hypertension. Albuminuria was detected in 32% of subjects.
Figure 1 shows the proportions of subjects with
good (⬍140/90 mm Hg), satisfactory (140–159/90–
94 mm Hg) and poor (⭓160/95 mm Hg) BP control
in those with and those without a diagnosis of
hypertension. Of all 260 subjects, 152 (58%) had
hypertension according to the WHO criteria, 49
(19%) of these had not previously been diagnosed as
being hypertensive. With respect to those previously
diagnosed as being hypertensive, one out of 10 had
good but one in two had poor BP control, and of
those not previously diagnosed, every fourth had
good and every third had poor BP control. Only
every fifth of all diabetic subjects was in good BP
control.
BP measurements had been registered at least
once in the previous 2 years in 233 (90%) subjects;
in 98% of those with a diagnosis of hypertension
and in 85% of those without this diagnosis. At least
two measurements each year had been carried out
in only 68% of subjects. In the subjects diagnosed
as being hypertensive, 61% were on monotherapy,
and 8% had three or more antihypertensive agents;
46% used angiotensin-converting enzyme (ACE)
inhibitors, 39% used beta-blockers, 31% used calcium antagonists and 31% used diuretics.
In Table 1, the study population (the eight subjects who had moved have been excluded) was
classified according to DBP under and equal or over
90 mm Hg in previous casual measurements. Those
with elevated DBP were younger, clearly more obese
and they had higher serum triglycerides and had a
higher prevalence of albuminuria (42% vs 27%).
Blood pressure control in type 2 diabetes
JA Hänninen et al
113
Figure 1 Blood pressure control according to a diagnosis of hypertension.
Table 1 Baseline characteristics and 5-year mortality of type 2
diabetic patients according to normo- and hypertension (diastolic
BP below or equal plus over 90 mm Hg). n = 252; the blood pressure of 26 subjects and albuminuria of additional 10 subjects had
not been measured
Number of
patients
(men/women)
Mean age (years)
(s.d.)
Median duration
of diabetes (years)
(range)
Mean blood
glucose (mmol/l)
(s.d.)
Mean HbA1c (%)
(s.d.)
Mean BMI (kg/m2)
(s.d.)
Mean total
cholesterol
(mmol/l) (s.d.)
Mean HDLcholesterol
(mmol/l) (s.d.)
Median
triglycerides
(mmol/l) (s.d.)
Albuminuria
⭓20 ␮g/min (%)
Hypertension (%)
Coronary heart
disease (%)
Deceased in 5
years (%)
Casual DBP
⬍90 mm Hg
Casual DBP
⭓90 mm Hg
P-value
116 (58/58)
110 (60/50)
0.494
58.0 (5.3)
55.1 (6.9)
⬍0.001
7.0 (0–33)
6.0 (0–26)
0.142
9.8 (2.8)
10.3 (2.8)
0.262
8.6 (2.2)
8.8 (1.9)
0.436
29.1 (5.1)
32.1 (5.5)
⬍0.001
5.8 (1.4)
6.1 (1.3)
0.060
1.1 (0.3)
1.1 (0.3)
0.681
1.7 (0.3–13.7)
2.3 (0.5–12.2)
30 (27)
43 (42)
0.018
47 (41)
29 (25)
55 (50)
26 (24)
0.152
0.811
9 (8)
10 (9)
0.763
⬍0.001
There were no differences in other parameters or in
5-year mortality between these two groups.
Of these 110 subjects with mean DBP ⭓90 mm Hg,
48 (44%) participated in the 24-h ABPM. The nonparticipants differed from the participants in that
they had a longer duration of diabetes (median 8 vs
5 years, P = 0.02, Wilcoxon), their gender (m/f 42/20
vs 18/30, P = 0.002 chi-square) and lower SBP (mean
149 vs 165 mm Hg, P = 0.007, t-test). There were no
differences in age, BMI, lipid values, UAER, initial
hypertension, coronary heart disease or 5-year mortality between the participant and non-participant
groups.
In Table 2, the results of the 24-h ABPM are
shown. The mean daytime BP was 21/6 mm Hg
lower in ABPM compared with casual measurements. White coat hypertension was diagnosed in
8/48 (17%) and non-dipper phenomenon in 18/47
(38%) of the hypertensive diabetic subjects. One of
the participants had both white coat hypertension
and non-dipper phenomenon. Those with white
coat hypertension had the lowest mean BP in 24-h
ABPM and their mean daytime BP was 33/18
mm Hg lower compared with casual measurements.
Of the non-dippers, 12/18 (67%) had higher MAP
during the night than during the daytime. One
ABPM recording failed to provide night-time BP
values. Two participants died during the follow-up,
and they both had sustained hypertension. No survival analyses were performed.
There were no differences in lipid values, BMI,
albuminuria or initial coronary heart-disease
between subjects with white coat and those with
sustained hypertension, and between non-dippers
and dippers. Those with white coat hypertension
tended to have a shorter duration of diabetes (P =
0.007, Wilcoxon) and less often had a diagnosis of
hypertension (P = 0.06, chi-square). On the contrary,
those with non-dipper phenomenon tended to have
a longer duration of diabetes (P = 0.02) and also were
more often likely to have been diagnosed as being
hypertensive (P = 0.07).
The calculation of statistical power to reveal significant differences in UAER in relation to the sample size is shown in Figure 2. With logarithmic
transformation of the UAER variable, the mean was
1.13, median 1.08, s.d. 0.62 and variance 0.38. It
Journal of Human Hypertension
Blood pressure control in type 2 diabetes
JA Hänninen et al
114
Table 2 BP values in casual and 24-h ABPM measurements of
hypertensive diabetic subjects, subgrouped according to white
coat hypertension and non-dipper phenomenon
All participants (n = 48)
Mean casual BP values
(mm Hg) (s.d.)
Mean ABPM day BP
(mm Hg) (s.d.)
Mean ABPM night BP
(mm Hg) (s.d.)
Mean casual BP values
(mm Hg) (s.d.)
Mean ABPM day BP
(mm Hg) (s.d.)
Mean ABPM night BP
(mm Hg) (s.d.)
Median age (years)
(range)
Diagnosed hypertension
(%)
Median duration of
diabetes (years) (range)
Mean BMI (kg/m2) (s.d.)
Median albuminuria
(␮g/min) (range)
Mean casual BP values
(mm Hg) (s.d.)
Mean ABPM day BP
(mm Hg) (s.d.)
Mean ABPM night BP
(mm Hg) (s.d.)
Median age (years)
(range)
Diagnosed hypertension
(%)
Median duration of
diabetes (years) (range)
Mean BMI (kg/m2) (s.d.)
Median albuminuria
(␮g/min) (range)
165/97 (17/5)
144/91 (13/9)
139/83 (18/11)
White coat
hypertension
(n = 8)
Sustained
hypertension
(n = 40)
162/98 (12/2)
166/97 (18/6)
129/80 (8/4)
147/93 (12/8)
122/72 (16/4)
142/85 (17/10)
58 (44 –63)
54 (39–64)
1 (13%)
20 (50)
2.5 (1–12)
5 (1–22)
33.1 (6.2)
14.5 (3–65)
32.3 (5.3)
9 (1–700)
Non-dipper
phenomenon
(n = 18)
Dipper
phenomenon
(n = 29)
172/98 (19/7)
161/97 (16/5)
145/89 (11/9)
144/93 (15/9)
152/91 (12/8)
131/78 (16/9)
55 (38–64)
58 (42–63)
10 (34)
11 (61)
4 (1–14)
7 (1–22)
31.7 (5.4)
10 (1–700)
32.8 (4.8)
10 (1–106)
could be estimated that these data could have
revealed a 8 ␮g/min UAER decrease in the subjects
with white coat hypertension and a 30 ␮g/min
UAER increase in those with the non-dipper
phenomenon with 90% degree of power and a 95%
confidence level.
Discussion
Hypertension was very common and BP control was
unsatisfactory in subjects with type 2 diabetes,
which is in line with previous studies.3,4 –6 Of the
hypertensive subjects, only 10% had good BP control and 50% had poor BP control. There may be
several reasons for this unsatisfactory outcome.
Firstly, elevated BP had not been detected, since
one-third of subjects with poor BP control did not
have a diagnosis of hypertension. In Finland, a mean
DBP ⬎95 mm Hg has been the criterion for 75%
Journal of Human Hypertension
Figure 2 Difference in UAER detected statistically by sample
size. The curves show upper and lower limits from 10 ␮g/min
with 90% power and 95% confidence level.
reimbursement (instead of 50%) of the costs of antihypertensive drugs for diabetic subjects from the
Social Security Institute, which may have influenced the initiation of antihypertensive agents. Secondly, antihypertensive treatment was inadequate,
since most hypertensive subjects were on monotherapy.2 Casual BP measurements had been performed
twice a year in two-thirds of the subjects, which is
far less than the recommended frequency.20
The other main result was that UAER can be used
as an indicator of BP control. Long-term poor BP
control in casual measurements was associated with
albuminuria, which agrees with the previous studies.12–14 In the present study, no differences in 5-year
mortality were found, which was probably due to
the small number of the study population and the
short follow-up time.22
The participation rate was low in the 24-h ABPM.
This measurement required two visits to the health
centre and the inconvienience of wearing the device
for 24 h. The non-participants were more often men
and those with mild hypertension, which may
underestimate the prevalence of white coat hypertension. The results showed that the compliance of
patients to participate in the ABPM was not very
high. Further, it has been shown that several ambulatory BP measurements are needed to get reliable
results.23 In the present study, the sample size was
too low to reveal the possible differences in the
between those with white coat hypertension and
those with the non-dipper phenomenon compared
with those with sustained hypertension.
In our study population, the prevalence of white
coat hypertension and the non-dipper phenomenon
were similar to the previous studies.16,18 Our results
suggest that white coat hypertension may represent
the first stage in the development of hypertension.
Two participants died during the follow-up, and no
conclusion can be drawn from the mortality.
It can be concluded that BP control was unsatisfactory in most subjects with type 2 diabetes. The
frequency of casual BP measurements could well be
Blood pressure control in type 2 diabetes
JA Hänninen et al
increased and the decision to initiate antihypertensive treatment should be based on these measurements. Education for doctors is needed if one wishes
to achieve earlier diagnosis of hypertension and
more aggressive antihypertensive treatment. Occurrence of microalbuminuria was associated with poor
BP control in long-term casual measurements and
UAER may be useful in assessing the BP control.
Further studies with greater sample sizes are needed
to assess the position of 24-h ABPM in care of hypertensive subjects with type 2 diabetes.
Acknowledgements
We thank nurses Sirpa Häkkinen and Merja Nykänen for their help in ABPM measurements and Mr
Paavo Mäkinen for skillful help in statistical analyses.
10
11
12
13
14
15
References
1 UK Prospective Diabetes Study Group. Tight blood
pressure control and risk of macrovascular and
microvascular complications in type 2 diabetes:
UKPDS 38. BMJ 1998; 317: 703–713.
2 Hansson L et al. (The HOT Study Group). Effects of
intensive blood-pressure lowering and low-dose
aspirin in patients with hypertension: principal results
of the Hypertension Optimal Treatment (HOT) randomised trial. Lancet 1998; 351: 1755–1762.
3 Uusitupa MIJ et al. Ten-year cardiovascular mortality
in relation to risk factors and abnormalities in lipoprotein composition in type 2 diabetic and non-diabetic
subjects. Diabetologia 1993; 36: 1175–1184.
4 Kumpusalo E, Pärnänen H, Takala J. Treatment situation of hypertensive patients in Finnish primary
health care. Blood Press 1997; 6 (Suppl 1): 35–39.
5 Du X et al. Case-control study of stroke and the quality
of hypertension care in north west England. BMJ 1997;
314: 272–276.
6 Mashru M, Lant A. Interpractice audit of diagnosis and
management of hypertension in primary care: educational intervention and review of medical records.
BMJ 1997; 314: 942–946.
7 Kemple TJ, Hayter SR. Audit of diabetes in general
practice. BMJ 1991; 302: 451– 453.
8 Hänninen J, Keinänen-Kiukaanniemi S, Takala J.
Population-based audit of non-insulin-dependent diabetic patients aged under 65 years in primary health
care. Scand J Prim Health Care 1998; 16: 227–232.
9 Niskanen LK, Penttilä I, Parviainen M, Uusitupa MI.
16
17
18
19
20
21
22
23
Evolution, risk factors, and prognostic implications of
albuminuria in NIDDM. Diabetes Care 1996; 19: 486–
493.
MacLeod JM, Lutale J, Marshall SM. Albumin
excretion and vascular deaths in NIDDM. Diabetologia
1995; 38: 610–616.
Gall MA et al. Albuminuria and poor glycemic control
predict mortality in NIDDM. Diabetes 1995; 44:
1303–1309.
Nielsen S et al. Albuminuria and 24-h ambulatory
blood pressure in normoalbuminuric and microalbuminuric NIDDM patients. Diabetes Care 1995; 18:
1434 –1441.
Mehler PS, Jeffers BW, Estacio R, Schrier RW. Associations of hypertension and complications in non-insulin-dependent diabetes mellitus. Am J Hypertens 1997;
10: 152–161.
Schmitz A, V{th M, Mogensen CE. Systolic blood
pressure relates to the rate of progression of albuminuria in NIDDM. Diabetologia 1994; 37: 1251–1258.
Høegholm A et al. Microalbuminuria in 411 untreated
individuals with established hypertension, white coat
hypertension and normotension. Hypertension 1994;
24: 101–105.
Nielsen FS et al. White coat hypertension in NIDDM
patients with and without incipient and overt diabetic
nephropathy. Diabetes Care 1997; 20: 859–863.
Fogari R et al. Urinary albumin excretion and nocturnal blood pressure in hypertensive patients with type
II diabetes mellitus. Am J Hypertens 1994; 7(9 Pt 1):
808–813.
Fogari R et al. Ambulatory blood pressure monitoring
in normotensive and hypertensive type 2 diabetics.
Prevalence of impaired blood pressure patterns. Am J
Hypertens 1993; 6: 1–7.
WHO Expert Committee. Arterial Hypertension. Technical Report Series No. 628. Geneva. World Health
Organisation, 1978.
European NIDDM Policy Group. Desktop Guide for the
Management of Non-Insulin-Dependent Diabetes Mellitus. (2nd edn.) Kirchheim: Mainz, 1993, p 8.
Tisler A, Barna I, de Chatel R. Accuracy testing of a
24-hour ambulatory blood pressure monitor using
standard zero and random zero sphygmomanometer.
Hungarian Med J 1994; 135: 1415–1419.
Hanefeld M et al. Risk factors for myocardial infarction
and death in newly detected NIDDM: the Diabetes
Intervention Study, 11 year follow-up. Diabetologia
1996; 39: 1477–1583.
Palatini P et al. White-coat hypertension: a selection
bias? Harvest Study Investigators. Hypertension and
Ambulatory Recording Venetia Study. J Hypertens
1998; 16: 977–984.
115
Journal of Human Hypertension