OBJECTIVE—To evaluate pulse pressure changes according to duration of type 1 diabetes and to assess the influence of posture.
RESEARCH DESIGN AND METHODS—We performed continuous measurement of blood pressure with a Finapres device during a 3 × 1 min posture test (standing, squatting, standing) in 159 type 1 diabetic patients divided into four groups according to diabetes duration (≤10, 11–20, 21–30, and >30 years, groups 1–4, respectively) and compared the results with those of age-matched nondiabetic subjects.
RESULTS—Pulse pressure progressively increased according to type 1 diabetes duration (P < 0.0001), especially in women, but not in age-matched nondiabetic subjects (NS). Pulse-pressure increase from group 1 to group 4 was amplified in the squatting position (from 50 ± 17 to 69 ± 14 mmHg) compared with standing (from 44 ± 15 to 55 ± 12 mmHg).
CONCLUSIONS—Pulse pressure increases according to type 1 diabetes duration more in women than in men, and the squatting position sensitizes such pulse-pressure increase in both sexes.
Early vascular aging was reported in type 1 diabetes (1–5). Arterial pulse pressure, a surrogate marker of artery stiffness, is an independent cardiovascular disease (CVD) risk factor in type 1 diabetic patients (6), as in nondiabetic (7–9) and type 2 diabetic (10,11) individuals. This cross-sectional controlled study investigated the influence of type 1 diabetes duration, sex, and posture on pulse pressure.
RESEARCH DESIGN AND METHODS—
Monitoring of blood pressure
Systolic and diastolic blood pressure were measured continuously with a Finapres (finger arterial pressure) instrument (Ohmeda) (13–15) during an orthostatic test (successively 1 min standing, 1 min squatting, 1 min standing) (16–18). Mean ± SD blood pressure and pulse pressure (systolic minus diastolic) were calculated throughout the test (overall), during initial standing only, and in the squatting position.
Statistical analysis
We performed univariate analysis using ANOVA followed by Student's t test, with two-tailed P values <0.05 considered as significant.
RESULTS—
Mean blood pressure remained almost stable across the four diabetic groups, whereas pulse pressure progressively increased according to type 1 diabetes duration (ANOVA P < 0.0001; Table 1). There was a marked difference in overall pulse pressure between groups 1 and 2 and groups 3 and 4 (49 ± 14 vs. 59 ± 14 mmHg; P < 0.00002), contrasting with no difference in age-matched control subjects (mean 35 years for groups 1 and 2 vs. 46 years for groups 3 and 4: 51 ± 12 vs. 50 ± 15 mmHg; NS).
Pulse pressure levels were lower in diabetic and nondiabetic women compared with men, except in group 4 with type 1 diabetes >30 years (data not shown). The progressive pulse pressure increase from group 1 to group 4 was almost double in women (from 44 ± 16 to 63 ± 11 mmHg) compared with men (from 51 ± 15 to 61 ± 13 mmHg) with type 1 diabetes.
The gradual rise in pulse pressure from group 1 to group 4 in type 1 diabetic patients, already present in the initial standing position (from 44 to 55 mmHg), was markedly amplified by squatting (from 50 to 69 mmHg) (Table 1).
CONCLUSIONS—
The present study essentially demonstrates: 1) a progressive pulse pressure increase according to type 1 diabetes duration, especially when it exceeds 20 years; 2) an early pulse pressure rise in type 1 diabetic subjects <50 years of age, especially in women; and 3) the amplification of this phenomenon by squatting.
In the Finnish Diabetic Nephropathy (FinnDiane) study, pulse pressure was higher in diabetic patients than in nondiabetic subjects of all age categories, including in individuals <45 years of age (5). However, previous observations showed no (19) or only marginally increased pulse pressure values (<5 mmHg) in young (25–30 years of age) type 1 diabetic patients (2,3,20). A recent study reported an 11-mmHg higher brachial pulse pressure and an 8-mmHg higher carotid pulse pressure in patients with 15.3 years of type 1 diabetes duration compared with control subjects (21). We observed substantial pulse pressure increase (difference of almost 10 mmHg) in type 1 diabetic patients aged >40 years with disease duration over 20 years compared with age-matched control subjects. Multiple regression analysis of the FinnDiane study indicates that type 1 diabetes duration per se has a considerable impact on pulse pressure, independent of age and renal involvement (5). Pulse pressure rises with age, more so in women (22). In healthy subjects, pulse pressure was higher in men than in women, although the difference decreased from <20 to 40–49 years of age (12); in the FinnDiane study, pulse pressure levels that were lower in nondiabetic women compared with men under the age of 40–45 years tended to become higher afterward. In the diabetic population, an earlier and steeper pulse pressure rise according to age was shown in women compared with men (5). Increased aortic wall stiffness with type 1 diabetes was more markedly related to diabetes duration in women than in men (23).
Squatting increases preload by augmentation of venous return and increases cardiac output without changes in systemic vascular resistance (16,17,24), resulting in greater increase of systolic than diastolic blood pressure and pulse pressure augmentation (15,25). Interestingly, the squatting-induced pulse pressure rise markedly increased with type 1 diabetes duration and almost doubled in patients with >20 years of type 1 diabetes compared those with <20 years of type 1 diabetes. Both increased arterial stiffness and alteration of wave reflection may contribute to the predominant pulse pressure increase in the squatting position with type 1 diabetes duration (26). Furthermore, AGEs may profoundly alter the histomorphometric arterial pattern, reducing compliance and distensibility, and NO-dependent vascular dilation (27).
Our study has some limitations. First, the cross-sectional design did not allow us to follow patients prospectively. Second, pulse pressure was measured at the finger site. However, pulse pressure changes in radial artery and finger were almost similar during a head-up tilt (28). Finally, type 1 diabetic patients were not well controlled, precluding any extrapolation to type 1 diabetic patients with excellent glucose control.
The Framingham Heart Study showed that pulse pressure is an excellent predictor of coronary heart disease risk, at least in nondiabetic individuals >50 years of age (29). Pulse pressure was also a major determinant of CVD complications in young type 1 diabetic individuals of the Eurodiab Prospective Complications study (6), in agreement with accelerated vascular aging and earlier arterial stiffening in type 1 diabetes (1,4,19,21). The greater pulse pressure increase in diabetic women according to disease duration (5) may contribute to the higher relative (vs. age-matched nondiabetic control subjects) CVD risk compared with diabetic men (30). AGEs were associated with increased pulse pressure in type 1 diabetes (27), an additional argument for intensifying glucose control to reduce CVD risk (31).
In conclusion, type 1 diabetes is associated with a progressive pulse pressure increase according to disease duration, a finding suggestive of accelerated arterial stiffening that may explain the higher CVD risk of type 1 diabetic patients, especially in women. The squatting position sensitizes the detection of increased pulse pressure in diabetic patients and may help detect patients at higher risk.
Characteristics of the type 1 diabetic patients and nondiabetic control subjects in the four subgroups divided according to diabetes duration and age matching, respectively, and mean blood pressure and pulse pressure levels recorded during the whole test (overall) in the initial standing position and in the squatting position
. | Group 1* . | . | Group 2* . | . | Group 3* . | . | Group 4* . | . | P† . | . | |||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Diabetic . | Nondiabetic . | Diabetic . | Nondiabetic . | Diabetic . | Nondiabetic . | Diabetic . | Nondiabetic . | Diabetic . | Nondiabetic . | |||||
n | 39 | 30 | 45 | 30 | 57 | 30 | 18 | 28 | |||||||
Age (years) | 34 ± 10 | 33 ± 10 | 37 ± 10 | 37 ± 11 | 45 ± 9 | 44 ± 11 | 48 ± 8 | 48 ± 9 | 0.0000 | 0.0000 | |||||
Sex (F/M) | 22/17 | 17/13 | 21/24 | 14/16 | 24/33 | 14/16 | 8/10 | 18/10 | |||||||
BMI (kg/m2) | 22.6 ± 3.3 | 23.1 ± 2.8 | 23.5 ± 2.4 | 23.6 ± 3.3 | 24.3 ± 4.4 | 24.7 ± 4.5 | 23.1 ± 2.2 | 24.4 ± 3.7 | 0.1005 | 0.3055 | |||||
Type 1 diabetes duration (years) | 6 ± 3 | — | 16 ± 3 | — | 26 ± 3 | — | 38 ± 8 | — | 0.0000 | NA | |||||
A1C (%) | 8.92 ± 1.68 | — | 8.72 ± 1.19 | — | 8.43 ± 1.25 | — | 8.85 ± 1.18 | — | 0.4215 | NA | |||||
Mean blood pressure (mmHg) | |||||||||||||||
Overall | 80 ± 10 | 82 ± 11 | 80 ± 11 | 80 ± 13 | 86 ± 12 | 85 ± 16 | 80 ± 13 | 84 ± 14 | 0.0149 | 0.1946 | |||||
Standing | 79 ± 10 | 81 ± 10 | 79 ± 12 | 79 ± 12 | 84 ± 12 | 84 ± 16 | 77 ± 13 | 83 ± 16 | 0.0496 | 0.1816 | |||||
Squatting | 84 ± 13 | 85 ± 12 | 87 ± 13 | 86 ± 15 | 94 ± 13 | 90 ± 16 | 87 ± 14 | 89 ± 15 | 0.0015 | 0.2422 | |||||
Pulse pressure (mmHg) | |||||||||||||||
Overall | 48 ± 16 | 49 ± 11 | 52 ± 13 | 53 ± 13 | 59 ± 14‡ | 49 ± 14 | 62 ± 12§ | 51 ± 15 | 0.0001 | 0.3783 | |||||
Standing | 44 ± 15 | 45 ± 10 | 47 ± 13 | 48 ± 12 | 54 ± 14 | 46 ± 15 | 55 ± 12 | 49 ± 15 | 0.0004 | 0.1885 | |||||
Squatting | 50 ± 17 | 51 ± 13 | 56 ± 15 | 55 ± 13 | 64 ± 16‖ | 51 ± 15 | 69 ± 14‖ | 52 ± 15 | 0.0000 | 0.4851 |
. | Group 1* . | . | Group 2* . | . | Group 3* . | . | Group 4* . | . | P† . | . | |||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Diabetic . | Nondiabetic . | Diabetic . | Nondiabetic . | Diabetic . | Nondiabetic . | Diabetic . | Nondiabetic . | Diabetic . | Nondiabetic . | |||||
n | 39 | 30 | 45 | 30 | 57 | 30 | 18 | 28 | |||||||
Age (years) | 34 ± 10 | 33 ± 10 | 37 ± 10 | 37 ± 11 | 45 ± 9 | 44 ± 11 | 48 ± 8 | 48 ± 9 | 0.0000 | 0.0000 | |||||
Sex (F/M) | 22/17 | 17/13 | 21/24 | 14/16 | 24/33 | 14/16 | 8/10 | 18/10 | |||||||
BMI (kg/m2) | 22.6 ± 3.3 | 23.1 ± 2.8 | 23.5 ± 2.4 | 23.6 ± 3.3 | 24.3 ± 4.4 | 24.7 ± 4.5 | 23.1 ± 2.2 | 24.4 ± 3.7 | 0.1005 | 0.3055 | |||||
Type 1 diabetes duration (years) | 6 ± 3 | — | 16 ± 3 | — | 26 ± 3 | — | 38 ± 8 | — | 0.0000 | NA | |||||
A1C (%) | 8.92 ± 1.68 | — | 8.72 ± 1.19 | — | 8.43 ± 1.25 | — | 8.85 ± 1.18 | — | 0.4215 | NA | |||||
Mean blood pressure (mmHg) | |||||||||||||||
Overall | 80 ± 10 | 82 ± 11 | 80 ± 11 | 80 ± 13 | 86 ± 12 | 85 ± 16 | 80 ± 13 | 84 ± 14 | 0.0149 | 0.1946 | |||||
Standing | 79 ± 10 | 81 ± 10 | 79 ± 12 | 79 ± 12 | 84 ± 12 | 84 ± 16 | 77 ± 13 | 83 ± 16 | 0.0496 | 0.1816 | |||||
Squatting | 84 ± 13 | 85 ± 12 | 87 ± 13 | 86 ± 15 | 94 ± 13 | 90 ± 16 | 87 ± 14 | 89 ± 15 | 0.0015 | 0.2422 | |||||
Pulse pressure (mmHg) | |||||||||||||||
Overall | 48 ± 16 | 49 ± 11 | 52 ± 13 | 53 ± 13 | 59 ± 14‡ | 49 ± 14 | 62 ± 12§ | 51 ± 15 | 0.0001 | 0.3783 | |||||
Standing | 44 ± 15 | 45 ± 10 | 47 ± 13 | 48 ± 12 | 54 ± 14 | 46 ± 15 | 55 ± 12 | 49 ± 15 | 0.0004 | 0.1885 | |||||
Squatting | 50 ± 17 | 51 ± 13 | 56 ± 15 | 55 ± 13 | 64 ± 16‖ | 51 ± 15 | 69 ± 14‖ | 52 ± 15 | 0.0000 | 0.4851 |
Data are means ± SD.
Group 1: diabetes duration ≤10 years, group 2: diabetes duration 11–20 years, group 3: diabetes duration 21–30 years, group 4: diabetes duration ≥30 years.
P value by ANOVA for groups 1–4. NA: Not applicable.
P < 0.05,
P < 0.01, and
P < 0.001, all for diabetic vs. nondiabetic.
Article Information
Parts of this study were presented in poster form at the 41st Meeting of the European Association for the Study of Diabetes, Athens, Greece, 10–15 September 2005, and as oral communication at the French Hypertension Meeting, Paris, France, 15–16 December 2005.
References
Published ahead of print at http://care.diabetesjournals.org on 31 October 2007. DOI: 10.2337/dc07-1646.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C Section 1734 solely to indicate this fact.