Association of Systolic and Diastolic Blood Pressure with the Risk of End-Stage Renal Disease in Older Type 2 Diabetes Mellitus Patients without Cardiovascular Disease: A Nationwide Population-Based Study
Article information
Abstract
Background
There is insufficient evidence to determine a precise blood pressure target in older adults with diabetes mellitus. In this study, we evaluated the potential relationship between blood pressure levels and end-stage renal disease (ESRD) in older type 2 diabetes mellitus (T2DM) patients without ESRD using a nationwide longitudinal population dataset.
Methods
We performed a retrospective, observational, cohort study including 267,156 older (≥65 years old) patients with T2DM and without ESRD from 2009 to 2018 based on the National Health Information Database. We divided the participants into eight groups based on their systolic blood pressure (SBP) and diastolic blood pressure (DBP). The primary outcome was ESRD. All outcomes were analyzed using Cox proportional hazards regression analysis while controlling for baseline covariates.
Results
During a median follow-up of 7.26 years, the incidence rate of ESRD was 2.03 per 1,000 person-years. In multivariable Cox proportional hazard modeling, the risk of the primary outcome was the lowest in groups with an SBP of 100–119 mm Hg and DBP of <80 mm Hg. In subgroup analysis according to the use of hypertension medication, there was a significant difference in DBP (P for interaction=0.026) but no difference in SBP (P for interaction=0.247). The risk of ESRD was the lowest in patients with an SBP of 110–129 mm Hg taking hypertension medication and the highest in the group with an SBP of ≥160 mm Hg.
Conclusion
Maintaining blood pressure at less than 120/80 mm Hg might prevent progression to ESRD in older T2DM patients without cardiovascular disease.
Highlights
• This nationwide cohort included 267,156 older adults with T2DM and no CVD.
• The lowest risk of ESRD was observed at SBP 111–120 mmHg and DBP <80 mmHg.
• ESRD risk increased sharply at SBP ≥160 mmHg or DBP ≥90 mmHg.
• Maintaining BP <120/80 mmHg reduced ESRD risk by ~27% compared with SBP <140 mmHg
• The findings were consistent regardless of antihypertensive medication use.
INTRODUCTION
In older patients with diabetes, hypertension is the most common comorbidity (60%–85%) [1-5]. Elevated blood pressure (BP) is a major risk factor for both macrovascular and microvascular complications [6,7]. There is significant evidence for the benefits of BP control in diabetes patients with hypertension in terms of reducing major macrovascular and microvascular complications of diabetes as well as reducing mortality [8-10]. Prospective observational studies in people without major illnesses at baseline have consistently shown that the association between BP and the risk of cardiovascular events is direct and continuous at BP levels of 115/75 mm Hg or higher [11].
Patients with type 2 diabetes mellitus (T2DM) typically have a high rate of cardiovascular morbidity and mortality, particularly older patients. In older diabetes patients, there is a significant positive association between elevated systolic blood pressure (SBP) and the risk of end-stage renal disease (ESRD) [11-14]. Hence, BP control is also expected to confer benefits in older people, and BP control becomes increasingly important in the prevention of ESRD in older people [12]. However, previous studies may have predominantly studied small and broader age populations, therefore BP goals in older diabetes patients have been extensively debated during the past decade [15]. There is little information regarding target BP levels in older patients with T2DM and hypertension, and there has been some discrepancy in the target BP levels among different guidelines for older people with diabetes [16-19].
In this study, we evaluated the potential relationships between different BP levels and ESRD events using a large-scale population dataset from the National Health Information Database (NHID). This research may provide unique insights into the age-related dynamics of the association between elevated BP and ESRD. And we adopt a longitudinal design, it may offer a more comprehensive understanding of the temporal relationship between elevated SBP and the development of ESRD in old diabetics. This could be a novel contribution to provide insights into the progression of renal disease over time.
METHODS
Study database
Data for our analysis were obtained from the NHID, a public database on health care utilization and health screening that contains sociodemographic and mortality information for the entire population of South Korea. The NHID contains data for the years 2002–2017. The NHID, established by the National Health Insurance Service, was launched in 2000 by integrating 375 insurance associations. The NHID provides longitudinal data for 97% of the Korean population, with linkage to the National Death Registry and National Health Screening Program [20,21]. The latter was initiated in 2009 and includes a medical interview and a postural examination, chest X-ray examination, blood test (including tests for fasting glucose and triglyceride levels), urine test, dental screening, and additional examinations. Approval for the study protocol was obtained from the Institutional Review Board of Hanyang University Guri Hospital (KBSMC 2018-01-036). The requirement for informed consent was waived by the board.
Study participants
This was a national observational cohort study that included 267,156 persons. Based on the NHID, 922,061 persons with diabetes participated in the National Health Screening Program in 2009. Among the 922,061 persons with diabetes, 642,042 individuals aged <65 years, 12,250 patients with a history of stroke, 6,843 patients with a history of myocadiac infarction, and 11,023 individuals lacking complete data were excluded from our study. Therefore, the total number of eligible participants was 249,903. Participants were categorized into eight groups based on SBP, diastolic blood pressure (DBP), and pulse pressure (PP).
Definitions of diabetes
Diabetes was defined based on the International Classification of Disease, 10th Revision (ICD-10) codes E11. Patients with T2DM were identified as those with at least one annual insurance claim for anti-diabetic medication under these ICD-10 codes or those with a fasting plasma glucose level ≥126 mg/dL from the National Health Screening Program.
Definitions of study outcome
The primary outcome was newly diagnosed ESRD, defined as: (1) Patients with ≥1 claim under ICD-10 codes N18–19 who received hemodialysis, peritoneal dialysis, or kidney transplantation. (2) Patients covered by the relieved Co-payment Policy for ESRD. The study population was followed from baseline to the onset of dialysis or until December 31, 2018, whichever came first.
Clinical and laboratory measurements
All participants completed a questionnaire on medical history, use of tobacco and alcohol, and exercise habits. Based on smoking habits, participants were categorized as non-smokers, ex-smokers, or current smokers. Based on alcohol consumption, participants were categorized as non-drinkers, moderate drinkers (<30 g/day), or heavy drinkers (≥30 g/day). Regular exercise was defined as moderate intensity exercise performed three or more times per week or vigorous intensity exercise performed five or more times per week. Body mass index (BMI) was calculated as the body weight (in kilograms) divided by height (in meters) squared. BP was measured using a standard procedure with a sphygmomanometer after the participant had rested for more than 5 minutes. Blood samples were collected after overnight fasting. Plasma glucose, total cholesterol, triglyceride, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol levels were measured. We calculated the glomerular filtration rate using the four-variable Modification of Diet in Renal Disease Study equation [22]. Baseline comorbidities were identified as dyslipidemia (ICD-10 code E78 with the use of lipid-lowering agents or a serum total cholesterol level ≥240 mg/dL) and chronic kidney disease (CKD; estimated glomerular filtration rate [eGFR] <60 mL/min/1.73 m2).
Data analyses
Baseline characteristics were analyzed using descriptive statistics. Categorical variables are described as the frequency and percentage. Continuous variables are described as the mean±standard deviation (SD) for normally distributed data and as the geometric mean and 95% confidence interval (CI) for non-normally distributed data. We compared the baseline characteristics of eight groups defined by SBP, DBP, and PP. Continuous variables were compared using one-way analysis of variance, while categorical variables were compared using the chi-square test. Data of the follow-up duration for each group were obtained. The incidence rate of ESRD was estimated for each group over the total follow-up period. Incidence curves were estimated using the Kaplan–Meier method, and the log-rank test was also performed. All outcomes were analyzed using Cox proportional hazards regression analysis while controlling for baseline covariates. A two-tailed P value less than 0.05 was considered significant. Analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA) and R program version 3.4.1 (The R Foundation for Statistical Computing, Vienna, Austria, http://www.R-project.org).
RESULTS
Baseline characteristics of the participants
Characteristics of participants in the eight SBP groups are described in Table 1. Participants in the higher SBP group were older, were more likely to be female, were heavy drinkers, were hypertension medication users, had a higher BMI and fasting plasma glucose level, and had a higher prevalence of dyslipidemia and CKD. Participants in the higher SBP group were less likely to be current smokers, regular exercisers, and insulin users. Similar patterns of baseline characteristics were noted in the eight DBP groups (Supplementary Table 1) and different reference SBP groups (Supplementary Table 2).
Baseline demographic and clinical characteristics according to systolic blood pressure groups
Systolic and diastolic blood pressure and end-stage renal disease
There were 4,153 (1.55%) events of the newly diagnosed ESRD over a 7.26-year mean follow-up period (Table 2). We selected SBP 121–130 mm Hg and DBP <70 mm Hg as the reference group because this group demonstrated the lower incidence rate of ESRD, thus providing a stable comparative baseline. The incidence rate of the newly diagnosed ESRD was 2.03 per 1,000 person-years. In patients with an SBP of 111–120 mm Hg, the incidence of the primary outcome was 1.50 per 1,000 person-years, which was 27% lower compared to the SBP 131–140 mm Hg group (relative risk [RR]; 0.85 [95% CI, 0.75 to 0.96] vs. 1.12 [95% CI, 1.02 to 1.24]) and lower than that observed in all other SBP groups (Table 2, Figs. 1A and 2A). In the multivariable Cox proportional hazard model with a reference SBP group (SBP 121–130 mm Hg), the risk of the primary outcome was significantly high with a higher SBP and an SBP of <100 mm Hg than that in the reference group after adjusting for age, sex, smoking status, alcohol consumption status, regular exercise, BMI, dyslipidemia, CKD, insulin treatment, number of oral diabetes medications, fasting plasma glucose level, and use of hypertension medication (Fig. 1A). In patients with a DBP of 76–80 mm Hg (as categorized clearly in Table 2), the incidence rate of the composite primary outcome was 1.83 per 1,000 person-years, lower than those in the other DBP groups (Table 2, Figs. 1B and 2B). When the DBP <70 mm Hg group was used as the reference group, the risk of the primary outcome increased significantly with higher DBP (DBP >90 mm Hg) in the multivariable Cox proportional hazard model adjusting for age, sex, smoking status, alcohol consumption status, regular exercise, BMI, dyslipidemia, CKD, insulin treatment, number of oral diabetes medications, fasting plasma glucose level, and use of hypertension medication (Fig. 1B).
Number of events, incidence rates, and hazard ratios of ESRD stratified by SBP and DBP levels
Hazard ratios (HRs) and 95% confidence intervals (CIs) for end-stage renal disease by (A) systolic blood pressure and (B) diastolic blood pressure. Adjusted for age, sex, smoking status, alcohol consumption status, regular exercise, body mass index, dyslipidemia, chronic kidney disease, insulin treatment, number of oral diabetes medications, fasting plasma glucose levels, and use of hypertension medication.
Kaplan–Meier estimates of survival and incidence probability with respect to end-stage renal disease (ESRD) in the eight groups defined by (A) systolic blood pressure (SBP) and (B) diastolic blood pressure (DBP) levels.
PP and ESRD
In patients with a PP of 45–50 mm Hg, the incidence of ESRD was 1.33 per 1,000 person-years, lower than those in the other PP groups (Supplementary Table 3). In the multivariable Cox proportional hazard model analysis, the risk of ESRD was higher in patients with T2DM and a PP >50 mm Hg than in the reference group (PP 46–50 mm Hg), after adjusting for age, sex, alcohol consumption status, smoking status, regular exercise, BMI, dyslipidemia, CKD, insulin treatment, number of oral diabetes medications, fasting plasma glucose level, and use of hypertension medication (Supplementary Figs. 1 and 2).
Subgroup analysis based on the use of antihypertensive medication
Subgroup analyses were performed for the risk of the composite primary outcome according to the use of hypertension medication (Fig. 3). There was no significant interaction between the subgroups of patients who used and did not use hypertension medication for the risk of the primary outcome according to SBP groups (P for interaction=0.69) (Fig. 3A). Furthermore, there was no interaction between the subgroups in terms of the risk of the primary outcome based on DBP groups (P for interaction=0.19) (Fig. 3B). There was no interaction between the subgroups for the risk of the primary outcome according to PP groups (P for interaction=0.851) (Supplementary Fig. 1).
Hazard ratios (HRs) and 95% confidence intervals (CIs) for end-stage renal disease by (A) systolic blood pressure and (B) diastolic blood pressure in subgroups treated with or without hypertension medication. Adjusted for age, sex, smoking status, alcohol consumption status, regular exercise, body mass index, dyslipidemia, chronic kidney disease, insulin treatment, number of oral diabetes medications, fasting plasma glucose levels, and use of hypertension medication.
DISCUSSION
In this national retrospective study, we examined the associations of SBP, DBP, and PP with the risk of ESRD in older diabetes patients without ESRD. During a median follow-up of 7.26 years, older diabetes patients without ESRD whose mean BP was less than 120/80 mm Hg had a significantly lower risk of ESRD. When SBP is lowered to less than 120 mm Hg, the risk of ESRD can be reduced by 27% compared to when managed less than 140 mm Hg (RR, 0.85 [95% CI, 0.75 to 0.96] vs. 1.12 [95% CI, 1.02 to 1.24]), and when DBP is lowered to less than 80 mm Hg, the risk of ESRD can be reduced by about 12%. The group with a mean PP of <50 mm Hg had the lowest risk of the primary outcome among all study participants. In the subgroup analysis according to the use of hypertension medications, similar results were found in participants using hypertension medications.
The meta-analysis of 19 trials was included to provide comparative context regarding the impact of intensive BP control in different studies, highlighting its mixed outcomes on renal endpoints. Intensive BP-lowering treatment achieved a reduction in the RR for albuminuria (10%; 95% CI, 3 to 16). However, intensive treatment had no clear effects on incidence of ESRD (RR, 10%; 95% CI, 6 to 23). In the subgroup analysis of those without CKD, intensive BP control significantly decreased the risk of CKD (RR, 16%; 95% CI, 7 to 24) [4].
The importance of strict BP control has been suggested based on the results of the Hypertension Optimal Treatment Trial and UK Prospective Diabetes Study [8,13]. A meta-analysis of 13 randomized controlled trials (RCTs) involving patients with diabetes or pre-diabetes showed that a reduction in SBP to 131–135 mm Hg reduced the risk of all-cause mortality by 13%, and more intensive SBP control (target: ≤130 mm Hg) was only associated with a greater reduction in the incidence of stroke [23]. Therefore, an SBP of ≤130 mm Hg has been recommended as a treatment goal in previous cardiovascular guidelines. However, the lack of a beneficial effect of strict SBP control (target: <120 mm Hg) compared with that of a conventional SBP target of 140 mm Hg, derived from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study, resulted in the modification of the target SBP in diabetes patients to 140 mm Hg in major clinical guidelines [5,24,25].
Our study findings support a more aggressive BP control target (<120/80 mm Hg) compared to the American Diabetes Association (ADA)’s recommended target of <130/80 mm Hg, indicating a potential benefit in further lowering the BP targets for older diabetic patients. This is a more intensive treatment goal than previous recommendations [26]. The European Society of Cardiology (ESC) recommends that in patients with diabetes, the target SBP should be ≤130 mm Hg but not <120 mm Hg. In older patients with diabetes (aged >65 years), the SBP target falls in the range of 130–139 mm Hg, and the target DBP is recommended to be <80 mm Hg but not <70 mm Hg [27,28].
In RCTs, the target SBP is 120–139 mm Hg, and it is recommended that the BP should not fall below the target BP; the ESC recommends a target SBP of 130–139 mm Hg, especially for older patients. The ADA also recommends the adjustment of the BP target to <130/80 mm Hg in healthy diabetes patients over 65 years of age and to 140/90 mm Hg in complex patients with poor health.
However, based on our national data, the rate of progression to ESRD below 120/80 mm Hg group was significantly low, and the same pattern was maintained in those with and without CKD. In this study, the prevalence of CKD (excluding ESRD) at baseline ranged from 22% to 28% depending on BP levels.
This study was not an RCT or meta-analysis, but it was a nationwide retrospective cohort study that analyzed the data of diabetes patients aged >65 years over a follow-up period of 7.26 years; therefore, our findings are important. In particular, ESRD is a representative complication of diabetes that causes great social and economic losses.
In the African American Study of Kidney Disease and Hypertension (AASK) trial, the renal outcome was different from that in the current study. The achieved BP averaged (SD) 128/78 (12/8) mm Hg in the lower BP group and 141/85 (12/7) mm Hg in the usual BP group. The mean (standard error [SE]) slope of the eGFR from baseline to 4 years did not differ significantly between the lower BP group (−2.21 [0.17] mL/min per 1.73 m2/year) and the usual BP group (−1.95 [0.17] mL/min per 1.73 m2/year; P=0.24) [29]. However, the duration of follow-up in the AASK trial was shorter than that in the current study (4 years vs. 7.6 years). In another longitudinal study (follow-up 9.8 years), the time-averaged BP was predictive of changes in serum creatinine concentrations [30]. To increase statistical power, long-term observation is required to demonstrate the effect of BP on the development of ESRD.
Discrepancies between our data and clinical guidelines may have been a result of ethnic differences. In most RCTs, the number of Asian participants is small, but our data were extracted from an Asian population.
In the Okinawa Dialysis Study registry, the RR of ESRD in those with high-normal BP and hypertension was significant for both men and women (reference group, BP <120/80 mm Hg). In diabetes patients, the adjusted RR of ESRD per 10-mm Hg increase in SBP was 1.21 (95% CI, 1.07 to 1.38; P=0.002) and per 10-mm Hg increase in DBP was 1.17 (95% CI, 0.93 to 1.46; P=0.16) in men. In women, the adjusted RR of ESRD per 10-mm Hg increase in SBP was 1.30 (95% CI, 1.12 to 1.50; P=0.0004) and per 10-mm Hg increase in DBP was 1.36 (95% CI, 1.02 to 1.81; P=0.03) [31]. These data are consistent with our findings. Therefore, control of BP within normal limits (<120/80 mm Hg) should be emphasized as a strategy for preventing ESRD in older diabetes patients.
This study had several limitations. First, the retrospective observational design of the study has inherent limitations. Although the analyses were adjusted for most available demographic and clinical variables, some unidentified variables could have affected the results. Second, we defined ESRD based on claims data; this may not be a completely accurate method for determining the number of cases. To overcome this problem, we defined the outcome according to the operational definition by combining diagnosis and prescription records. Third, this was not a prospective study; therefore, causality cannot be determined. However, to minimize the possible effects of reverse causality, participants with pre-existing ESRD were excluded. And last, we adjusted for CKD status to account for underlying renal function. However, we recognize that baseline eGFR is a critical predictor of progression to ESRD. Due to limitations in data availability, we were unable to include baseline eGFR in our multivariable model. This omission may lead to residual confounding, as variations in baseline kidney function could independently influence ESRD risk beyond what is captured by CKD status alone. Future research should incorporate baseline eGFR to enhance the precision of risk estimates associated with different SBP categories
The strengths of our study are that we used a large-scale nationwide database representing the entire Korean population. Second, we conducted analyses adjusted for available confounding cardiovascular risk factors.
In conclusion, adequate control of BP should be emphasized as a primary way to prevent ESRD, which means that a controlled study analyzing multiple risk factors for hypertension is needed. Based on the results of this nationwide retrospective study, older T2DM patients should be managed such that their BP is maintained at <120/80 mm Hg, which might prevent progression to ESRD in older T2DM patients without CVD.
SUPPLEMENTARY MATERIALS
Supplementary materials related to this article can be found online at https://doi.org/10.4093/dmj.2023.0364.
Baseline demographic and clinical characteristics according to diastolic blood pressure groups
Number of events, incidence rates, and hazard ratios of ESRD stratified by SBP levels
Number, incidence rate, and hazard ratio of end stage renal disease and all-cause mortality stratified by pulse pressure
Hazard ratios (HRs) and 95% confidence intervals (CIs) of end stage renal disease by pulse pressure according to (A) regardless of antihypertensive medication use and (B) the use of antihypertensive medications. Adjusted for age, sex, smoke, alcohol consumption, regular exercise, body mass index, dyslipidemia, chronic kidney disease, insulin treatment, number of oral diabetes medication, fasting plasma glucose level, and hypertension medication.
Kaplan–Meier estimates of survival and incidence probability by eight groups of pulse pressure (PP) for end-stage renal disease (ESRD).
Notes
CONFLICTS OF INTEREST
Sung Hoon Yu has been a associate editor of the Diabetes & Metabolism Journal since 2022. He was not involved in the review process of this article. Otherwise, there was no conflict of interest.
AUTHOR CONTRIBUTIONS
Conception or design: all author.
Acquisition, analysis, or interpretation of data: S.H., K.H., J.H.P.
Drafting the work or revising: S.H., S.H.Y.
Final approval of the manuscript: S.H., S.H.Y.
FUNDING
This work was supported by the research fund of Hanyang University (HY-2016).
ACKNOWLEDGMENTS
The National Health Information Database is maintained by the National Health Insurance Service of Korea. The authors would like to thank the National Health Insurance Service for their cooperation.