The prospective Kailuan Study showed that the exposure to elevated blood pressure accumulated over 3 examinations performed in 5 years was significantly associated with an increased incidence of diabetes, within the subgroup of individuals without previously impaired glucose tolerance and within the subgroup of individuals with previously impaired glucose tolerance. In a parallel manner, a higher accumulated exposure to elevated blood pressure was correlated with a higher rate of newly developed impairment of glucose tolerance (Additional file 2: Table S2; Tables 1, 2 and 3). For each increase in the cumSBP by 10 mmHg/year the risk of incident diabetes and of incident impairment of glucose tolerance increased by 3%. In multivariate analysis, this increase was independent of other diabetes-related related risk factors such as older age, gender, smoking, alcohol consumption, amount of physical exercise, level education and taking blood pressure-lowering drugs, and heart rate, body mass index, and fasting serum concentrations of glucose, triglycerides, high-sensitive C-reactive protein, high-density lipoproteins and low-density lipoproteins and uric acid.
The findings obtained in our study were in agreement with the results found in previous investigations. In a study Emdin and colleagues on more than 4 million adults, an increase in systolic blood pressure by 20 mmHg was correlated with a 58% higher risk of developing diabetes, and an increase in diastolic blood pressure by 10 mmHg was correlated with a 52% higher chance to develop diabetes [7]. The association between blood pressure and the incidence of diabetes mellitus (type 2) was also examined by Hayashi and colleagues [5] In a prospective study on more than 7000 Japanese men with an age ranging between 35 years and 60 years, high normal blood pressure as compared with normal blood pressure was associated with a 39% higher risk, and arterial hypertension was associated with a 76% higher risk, to develop diabetes mellitus. In the Candesartan Antihypertensive Survival Evaluation in the Japan trial on high-risk Japanese hypertensive patients, Yasuno and associates found that for each standard deviation increase in pulse pressure the hazard ratio for the development of diabetes was 1.44 in a multiple regression analysis [6]. A longitudinal investigation performed by Stahl and coworkers included more than 7000 men who were free of diabetes and who were followed-up for 35 years [16]. The hazard ratio for the incidence of diabetes was 1.43, 1.43 and 1.95 for individuals with a systolic blood pressure of 130–139 mmHg, 140–159 mmHg, and ≥160 mmHg, respectively, while a systolic blood pressure of less than 130 mmHg served as reference level. Analyzing the data of the Atherosclerosis Risk in Communities study, the Coronary Artery Risk Development in Young Adults study and the Framingham Heart Study offspring cohort together, Wei and colleagues found that the age-adjusted incidence of type 2 diabetes was increasingly higher across increasing blood pressure groups [17]. In contrast to the studies mentioned above, blood pressure factors was not predictive for incident diabetes in a study by Norberg and associates, whose investigation was a case-referent study nested within a population-based health survey [8]. In a similar manner in the investigation by Wang and colleagues, systolic and diastolic blood pressure did not predict the onset of diabetes [9].
Our study adds to the current knowledge on the association between hypertension and the development of diabetes and an impairment of glucose tolerance by taking into account the time-cumulative exposure to an increased blood pressure over a relatively long time of 5 years as baseline values, and by also addressing the development of an impairment of glucose tolerance. Studies on the associations between the time-cumulative exposure to risk factors for the incidence of major diseases have so far been performed with respect to the cumulative data on hyperglycemia and diabetic complications, the cumulative exposure to high serum concentrations of cholesterol and the incidence of coronary heart disease, and to the cumulative blood pressure exposure and the development of kidney damage [10,11,12].
The reasons for the association between time-cumulated blood pressure exposure and incident impairment of glucose tolerance and incident diabetes mellitus have remained elusive so far. Previous studies have shown that a low-grade inflammatory process occurs in both diabetes and hypertension both of which could be considered as chronic inflammatory diseases [18,19,20,21]. Correspondingly, inflammatory markers such as C-reactive protein are increased in patients with diabetes and in patients with arterial hypertension and do also predict the development of these diseases [22, 23]. Gene regulatory network analysis has demonstrated that oxidative stress is important for diabetes and hypertension, with the oxidative stress-mediated regulation cascade being the common mechanistic link among the pathogenesis of diabetes, arterial hypertension, and other related inflammatory diseases [24]. Also, insulin as a pleiotropic hormone plays a pivotal role in the development of arterial hypertension and diabetes.
In view of the pronounced increase in the prevalence of diabetes and arterial hypertension, the search of new screening tools that are economic and easily available in the clinical practice is of high practical importance. To date, several markers have been available to indicate an increased risk for the onset and progression of diabetes, such as inflammatory markers or proinsulin/insulin ratio, however they are not easily feasible for large scale screening [25, 26]. Furthermore, early identification of diabetes may be useful in terms of lifestyle intervention and prevention of long term complications. The results of the present study suggest that the time-cumulated exposure to elevated blood pressure is an additional risk factor predicting an increased likelihood to develop an impaired glucose tolerance and diabetes. Time-cumulated exposure to elevated blood pressure in combination with other parameters such as inflammatory markers, the proinsulin/insulin ratio, body mass index and amount of physical activity may be taken in a synopsis to estimate the risk of eventual onset of diabetes.
Limitations of our study should be discussed. First, the follow-up period of 2.22 years after the baseline period of 5 years was relatively short. With a longer follow-up and a higher number of patients developing diabetes or an impairment of glucose tolerance, the statistical significance of the associations between hypertension and the incidence might have become even clearer. Second, it has remained elusive, whether the multivariate analysis with adjustment for a multitude of parameters (including gender, age, smoking, alcohol consumption, amount of physical exercises, level education, taking of blood pressure-lowering drugs, heart rate, body mass index and fasting serum concentrations of glucose, triglycerides, high-sensitive C-reactive protein, high-density lipoproteins and low-density lipoproteins and uric acid) had not overlooked an additional factor potentially confounding the associations between hypertension and incidence of diabetes and impairment of glucose tolerance. In view of the robust statistical significance of these associations however, it might have been unlikely that a factor besides those already taken into account might have markedly impeded the associations between hypertension and incidence of diabetes and impairment of glucose tolerance as found in our study.