This study evaluated whether exogenous insulin use to control blood glucose in patients with type 2 diabetes was associated with the development of hypertension by analyzing data obtained from a large national sample in Taiwan. Exogenous insulin use was shown to be a significant risk factor for hypertension development.
Increased insulin levels are an early indication of insulin resistance. They have also been linked to the development of atherosclerotic diseases1 and hypertension.2 In patients with type 2 diabetes mellitus, exogenous insulin injections are sometimes necessary to obtain better control of blood glucose levels during the later stages of diabetes, when pancreatic failure occurs. Because insulin acts like a growth factor in the vasculature3 and produces a range of hemodynamic effects4-7 that promote atherosclerosis and hypertension development, the effects of exogenous insulin on hypertension risk are clinically important for the large number of patients with diabetes. Therefore, the current study evaluated the relationship between hypertension development and the use of exogenous insulin for blood glucose control among patients with type 2 diabetes, a sampling of which was obtained from the National Health Insurance database in Taiwan.
Subjects and methods
The procedure used to obtain a national sample from the Taiwanese National Health Insurance database, which covers more than 96% of the nation’s population, has been previously reported.8,9 A total of 256,036 patients with diabetes from 66 clinics and hospitals were identified through the database, taking into account all types of medical settings and a homogeneous geographical distribution. Anticipating a response rate of 70%, 128,572 (every 1 of 2 identified patients) were randomly chosen to obtain a sample of 90,000 subjects.
A questionnaire was given to subjects via the telephone from March 1, 1995, to April 30, 2002. The information extracted for analysis included weight, height, sex, age, usual systolic blood pressure, usual diastolic blood pressure, personal history of hypertension, parental history of hypertension, smoking history, year diabetes was diagnosed, symptoms at onset of diabetes, and method of treatment for differentiation between the different types of diabetes. For subjects who were taking insulin for blood glucose control, information regarding the time of treatment initiation was obtained. For those subjects who had a history of hypertension, the time of diagnosis was obtained. Subjects were considered to have hypertension if they had a history of hypertension or if they had a systolic blood pressure ≥ 140 mm Hg or a diastolic blood pressure ≥ 90 mm Hg, or both, with no history of hypertension.
Nearly 73% of subjects (n = 93,484) completed the questionnaire. A total of 3528 subjects were excluded because they had type 1 diabetes, 99 subjects were excluded because they were younger than 18 years of age, and 2007 subjects were excluded because they did not report their normal blood pressure, leaving 87,850 subjects. Of these, 5927 were using insulin injection to control blood glucose levels.
Student’s test for continuous variables and chi-square test for categorical variables were used to compare the differences in the baseline characteristics between users and nonusers of insulin. Insulin users were categorized into 3 groups based on how long they had been taking insulin: < 5 years, 5 to 9 years, and ≥ 10 years. The prevalence of hypertension among users and nonusers of insulin with regard to the duration of insulin use as well as the proportion of patients having a systolic blood pressure ≥ 140 mm Hg and a diastolic blood pressure ≥ 90 mm Hg with regard to insulin use were analyzed using the trend test and the chi-square test.
The adjusted odds ratios and 95% confidence intervals were estimated for hypertension using logistic regression analysis. Analyses were also performed separately for insulin users without hypertension at the time insulin use was initiated to ensure correct temporality of insulin use before hypertension developed. Using a Cox model, the relative risk of developing hypertension was compared between subjects who used insulin but did not have hypertension at the start of insulin use and subjects who did not use insulin and were free from hypertension at the time of diabetes diagnosis. The age and duration of diabetes entered into the Cox model were computed to the respective values at the beginning of follow-up. Parental hypertension, smoking, duration of diabetes, body mass index (BMI), sex, and age were the other independent variables entered into the logistic and Cox models simultaneously for adjustment.
The comparison of the baseline characteristics between users and nonusers of insulin showed that all variables differed significantly ( < .001), except for parental hypertension (19.9 vs 18.9; > .1). Insulin users had a higher prevalence of hypertension (61.3% vs 53.9%), had higher systolic blood pressures (135.3 ± 15.6 vs 132.9 ±12.9 mm Hg), had higher diastolic blood pressures (80.5 ± 9.3 vs 79.1 ± 8.3 mm Hg), were less often smokers (27.9% vs 30.7%), had a slightly lower BMI (24.4 ± 3.8 vs 24.6 ± 3.6 kg/m2), had diabetes longer (14.4 ± 7.5 vs 6.6 ± 6.1 years), were more likely to be women (60.4% vs 53.6%), and were older (63.1 ± 11.0 vs 62.2 ± 11.3 years).
shows the prevalence of hypertension for insulin users and nonusers in all subjects and in subjects using insulin but without hypertension at the start of insulin use. The adjusted odds ratios with regard to insulin use in all subjects and in subjects using insulin but without hypertension at the start of insulin use estimated from the logistic models are also shown at the top of Figure 1. In the Cox model, insulin users had a 1.45-fold greater risk of developing hypertension than nonusers.
shows the percentage distributions of subjects with systolic blood pressure ≥ 140 mm Hg and diastolic blood pressure ≥ 90 mm Hg for all subjects and for subjects using insulin but without hypertension at the start of insulin use.
As shown by the logistic regression models performed in all subjects, the risk of hypertension increased approximately 1.5-fold in subjects using insulin for > 10 years compared with nonusers (Figure 1). The risk increased approximately 2.2-fold when comparing subjects using insulin ≥ 10 years with those using insulin < 5 years in patients using insulin but without hypertension at the start of insulin use (Figure 1). In addition, a consistent dose-response relationship existed between the duration of insulin use and hypertension prevalence (Figure 1), blood pressure control status (Figure 2), and the odds ratios (Figure 1). The Cox proportional hazards model showed the correctness in temporality between the cause of insulin use and the effect of hypertension development.
The high concentrations of blood insulin caused by repetitive use of exogenous insulin may produce negative effects on the vascular system, resulting in increased blood pressure. The potential mechanisms could be explained based on the effects of insulin on sodium retention,10 ion transport,11 stimulation of sympathetic nerves,12 and induction of oxidative stress.13 However, insulin per se is also vasoactive in the peripheral vasculature and can be atherogenic because of its growth factor—like action.3
An animal study showed that a deficiency of insulin may prevent atherosclerosis, but this benefit was lost with insulin use.14 Insulin exerts a vasodilatory effect in the microvasculature of the skeletal muscle, which is important for insulin-mediated glucose uptake.4,5 However, in healthy subjects, insulin appears to produce a different effect on larger arteries and increases blood pressure. A marked increase in systolic blood pressure at either a low rate (1.5 mU/kg/min) or a high rate (15 mU/kg/min) of insulin infusion was shown in 1 study,6 providing evidence for the development of hypertension provoked by insulin. Another study showed that exogenous insulin infusion in healthy subjects could cause endothelial dysfunction and could abrogate endothelium dependent vasodilatation in the large conduit arteries, most likely by the induction of oxidative stress.7 The effects of insulin on the microvasculature in the skeletal muscle are reduced in patients with insulin resistance, such as in those with type 2 diabetes mellitus.4,5 The effects of excess exogenous insulin can be transferred from the nutritive to the nonnutritive vessels, such as the large conduit arteries, resulting in hypertension development.
This study assessed whether exogenous insulin use for blood glucose control was associated with hypertension in subjects with type 2 diabetes. Exogenous insulin use was shown to be a significant risk factor for the development of hypertension in these subjects.