ABSTRACT
Introduction
The optimal blood pressure (BP) targets for patients with type 2 diabetes (T2D) remain debated among international guidelines. This retrospective cohort analysis evaluated the associations of specific systolic and diastolic BP (SBP and DBP) categories with long‐term mortality in patients with T2D.
Materials and Methods
We analyzed 2,198 adults with T2D from the National Health and Nutrition Examination Survey (NHANES) 1999–2018. Participants were stratified by baseline SBP (<130, 130–140, ≥140 mmHg) and DBP (<80, 80–89, ≥90 mmHg). Multivariable Cox proportional hazards models were constructed to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for all‐cause, cardiovascular, cerebrovascular, and diabetes‐related mortality, adjusting for comprehensive baseline covariates.
Results
Over a median follow‐up of 97.0 months, an SBP ≥140 mmHg was significantly associated with elevated risks for all‐cause (HR = 1.66, 95% CI: 1.28–2.16), cardiovascular (HR = 1.53, 95% CI: 1.10–2.13), cerebrovascular (HR = 3.98, 95% CI: 1.71–9.27), and diabetes‐related mortality (HR = 1.84, 95% CI: 1.06–3.16) compared to an SBP <130 mmHg. Conversely, a DBP of 80–89 mmHg was associated with significantly lower all‐cause (HR = 0.67, 95% CI: 0.46–0.98) and cardiovascular mortality (HR = 0.57, 95% CI: 0.35–0.93) than a DBP <80 mmHg, indicating a distinct J‐curve phenomenon.
Conclusions
In this observational study, baseline SBP <140 mmHg and DBP 80–89 mmHg were associated with favorable survival outcomes in patients with T2D. These findings suggest that individualized blood pressure management warrants further investigation, taking into account the limitations of observational data and single baseline measurements.
Keywords: Blood pressure, Hypertension, Mortality
In this retrospective cohort analysis, systolic blood pressure ≥140 mmHg (vs <130) significantly increased all‐cause, cardiovascular, cerebrovascular, and diabetes mortality, whereas diastolic pressure 80–90 mmHg (vs <80) was associated with lower all‐cause and cardiovascular mortality.
Abbreviations
- ACC/AHA
American College of Cardiology/American Heart Association
- ACCORD
action to control cardiovascular risk in diabetes
- aHR
adjusted hazard ratio
- BMI
body mass index
- BP
blood pressure
- BPROAD
Blood Pressure Control in Patients with Type 2 Diabetes
- CAD
coronary artery disease
- CI
confidence interval
- CKD
chronic kidney disease
- CVD
cardiovascular disease
- DBP
diastolic blood pressure
- ESC
European Society of Cardiology
- ESH
European Society of Hypertension
- FPG
fasting plasma glucose
- HbA1c
glycated hemoglobin
- HDL‐C
high‐density lipoprotein cholesterol
- HR
hazard ratio
- ICD‐10
International Classification of Diseases, 10th Revision
- IQR
interquartile range
- LDL‐C
low‐density lipoprotein cholesterol
- LVAD
left ventricular assist device
- NCHS
National Center for Health Statistics
- NHANES
National Health and Nutrition Examination Survey
- OGTT
oral glucose tolerance test
- SBP
systolic blood pressure
- SGLT2
sodium‐glucose cotransporter 2
- SPRINT
Systolic Blood Pressure Intervention Trial
- T2D
type 2 diabetes
- TC
total cholesterol
- TG
triglycerides
- VADT
Veterans Affairs Diabetes Trial
- WC
waist circumference
INTRODUCTION
The coexistence of type 2 diabetes (T2D) and hypertension represents a pervasive and devastating clinical syndrome that synergistically accelerates the global burden of both macrovascular and microvascular complications 1 , 2 , 3 . Epidemiological data indicate that over 80% of adults with T2D are concurrently afflicted by hypertension, a combination that more than doubles the long‐term risk of cardiovascular disease (CVD), chronic kidney disease (CKD), and end‐stage renal disease compared to either condition in isolation 4 , 5 . The pathophysiological interplay between hyperglycemia and elevated blood pressure is bidirectional and complex; hypertension serves as a paramount driver of diabetic kidney disease, relentlessly promoting the progression of albuminuria, inducing profound endothelial dysfunction, and accelerating the decline of the glomerular filtration rate 6 , 7 .
Despite the universally recognized importance of stringent blood pressure (BP) control in mitigating these cardiometabolic sequelae, a profound and clinically significant divergence currently exists between the most authoritative international guidelines regarding the optimal BP targets for this exceptionally high‐risk population. The 2025 American College of Cardiology/American Heart Association (ACC/AHA) guidelines maintain an aggressive stance, advocating a Class 1 recommendation for a systolic blood pressure (SBP) target of <130 mmHg, with explicit encouragement to safely achieve <120 mmHg whenever tolerated, aiming to maximize the reduction of cardiovascular morbidity and mortality 8 . This intensive paradigm is primarily buttressed by recent milestone trials, such as the Blood Pressure Control in Patients with Type 2 Diabetes (BPROAD) trial, which demonstrated tangible cardiovascular benefits from intensive SBP control in selected diabetic cohorts 9 .
In stark contrast, the 2023 European Society of Hypertension (ESH) and European Society of Cardiology (ESC) guidelines arrive at a markedly more conservative conclusion 10 . While they similarly endorse a general treatment goal of <130/80 mmHg based on evidence of incremental protection—particularly concerning stroke prevention 11 —they explicitly advise against intensifying pharmacological treatment to deliberately achieve an SBP <120 mmHg or a diastolic blood pressure (DBP) <70 mmHg. This caution is deeply rooted in the nuanced legacy of trials such as the Action to Control Cardiovascular Risk in Diabetes (ACCORD) and Veterans Affairs Diabetes Trial (VADT), which, despite their complex designs, did not demonstrate a definitive mortality benefit for intensive BP control in diabetes and persistently raised concerns regarding the potential for iatrogenic harm from excessively low BP 12 , 13 , 14 .
This ongoing conflict between major cardiology societies creates a pervasive dilemma for practising clinicians. Furthermore, the complexity of T2D extends beyond singular glycemic or hemodynamic metrics. Modern understanding emphasizes a holistic cardiometabolic profile, where intricate systemic interactions—such as the cardiometabolic index—have been profoundly linked to severe neurological and psychiatric outcomes, including major depressive disorder, with stroke and diabetes acting as critical mediators 15 . Moreover, the optimal DBP range remains intensely contentious. Emerging observational evidence suggests a pronounced J‐shaped relationship, particularly in diabetic patients with established or subclinical coronary artery disease, where excessively low DBP may critically compromise myocardial perfusion during diastole 16 , 17 .
Therefore, this retrospective investigation leverages extensive longitudinal data from the National Health and Nutrition Examination Survey (NHANES) to rigorously evaluate the real‐world epidemiological associations of specific SBP and DBP categories with long‐term all‐cause and cause‐specific mortality in adults with T2D. Our study aims to provide robust observational evidence to contextualize these conflicting international recommendations and inform personalized clinical practice.
MATERIALS AND METHODS
Study design and setting
This retrospective cohort analysis utilized publicly available data from 10 cycles of the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2018. The NHANES employs a complex, multistage, probability sampling design to assess the health and nutritional status of adults and children in the United States. Before enrolling, all participants provided signed consent forms, indicating their informed agreement to take part in the study.
Participants
We systematically screened the NHANES database for eligible adults diagnosed with T2D. T2D was strictly defined by the presence of any of the following criteria: a glycated hemoglobin (HbA1c) level ≥6.5%, fasting plasma glucose (FPG) >7.0 mmol/L, an oral glucose tolerance test (OGTT) result >11.1 mmol/L, current use of antidiabetic medications, or a self‐reported physician diagnosis of type 2 diabetes.
The exclusion criteria were applied hierarchically as follows: the total number of participants is 102,956; (1) individuals aged <18 years, those with unavailable basic data, missing follow‐up time, or lacking survival status/cause of death data (n = 52,159); (2) participants lacking data on essential covariates (n = 34,580); and (3) participants missing the necessary indicators to confirm a diabetes diagnosis (questionnaire data, medicine use, HbA1c, FPG, or OGTT) (n = 14,019). Following these exclusions, a final analytic cohort of 2,198 individuals with T2D was established. The detailed participant selection process is visually represented in Figure 1.
Figure 1.
Flowchart of study participant selection. The diagram details the inclusion and exclusion criteria applied to identify the final analytic cohort of adults with type 2 diabetes from the National Health and Nutrition Examination Survey (NHANES, 1999–2018).
Variables and measurements
Exposures
The primary exposures were SBP and DBP. Following a mandatory resting period of 5 min in a seated position, blood pressure was meticulously assessed by trained physicians. To minimize measurement error, the mean systolic and diastolic BP readings utilized in our primary analysis were determined from the average of the first three consecutive measurements obtained at a single examination visit. The participants were categorized into three distinct groups according to their SBP (<130, 130–139, ≥140 mmHg) and DBP (<80, 80–89, ≥90 mmHg). These specific cut‐offs were selected a priori based on contemporary clinical guidelines.
Outcomes
The primary endpoints were all‐cause mortality, cardiovascular mortality, cerebrovascular mortality, and diabetes‐related mortality. Diabetes‐related mortality was explicitly defined as mortality with diabetes mellitus as the underlying cause of death (Code 007 of the NCHS Linked Mortality Files). The survival status and follow‐up duration of participants were ascertained via the NHANES public‐use mortality file, which is linked to the National Death Index utilizing a probability matching algorithm. The underlying causes of death were definitively coded according to the International Classification of Diseases, 10th Revision (ICD‐10) system.
Covariates
Demographic information (age, gender, race, education level, and income‐to‐poverty ratio) was gathered via standardized survey instruments. Health behaviors, such as daily alcohol intake (g/day), were collected through comprehensive questionnaires. Anthropometric measures, including body mass index (BMI) and waist circumference (WC), were recorded. Laboratory assays, including total cholesterol (TC), triglycerides (TG), high‐density lipoprotein cholesterol (HDL‐C), and low‐density lipoprotein cholesterol (LDL‐C), were conducted following defined techniques.
Statistical analysis
All statistical analyses were performed using R software (version 4.3.1). Continuous variables exhibiting skewed distributions were uniformly expressed as medians with interquartile ranges (IQRs), while categorical variables were reported as counts and percentages. Baseline characteristics across the defined blood pressure categories were compared utilizing the nonparametric Kruskal–Wallis test for continuous variables and the chi‐square test for categorical variables.
To evaluate the independent association between blood pressure categories and mortality outcomes, we constructed multivariable Cox proportional hazards regression models to calculate hazard ratios (HRs) and 95% confidence intervals (CIs). We presented both a crude (unadjusted) model and a fully adjusted model. The adjusted model comprehensively accounted for potential confounders, including the counterpart blood pressure metric (DBP or SBP), sex, age, race, education, income, alcohol intake, BMI, WC, TC, TG, HDL‐C, and LDL‐C. Cumulative event rates across blood pressure strata were visually assessed using Kaplan–Meier survival curves, and statistical comparisons were conducted using the log‐rank test. Finally, exploratory subgroup analyses stratified by age (<60 vs ≥60 years), sex, and BMI (<24, 24–28, >28 kg/m2) were performed to assess potential effect modification, evaluated via interaction terms (P for interaction). All statistical tests were two‐sided, and a P value <0.05 was considered statistically significant.
RESULTS
Baseline characteristics of the study population
A total of 2,198 participants diagnosed with T2D from the nationally representative NHANES (1999–2018) cohort were included in the final retrospective survival analysis. Over a median follow‐up duration of 97.0 months, we documented a robust number of endpoints, including 188 cases of cardiovascular mortality, 37 cases of cerebrovascular mortality, and 75 cases of mortality where diabetes was identified as the underlying cause.
The baseline demographic, clinical, and laboratory characteristics of the NHANES cohort, meticulously stratified by SBP categories (<130, 130–139, ≥140 mmHg), are detailed in Table 1. Participants in the highest SBP group (≥140 mmHg) were significantly older (median age 64.0 years vs 58.0 years in the <130 mmHg group; P < 0.001) and exhibited larger waist circumferences (P = 0.064). Furthermore, the racial distribution varied significantly, with the proportion of non‐Hispanic Black individuals increasing across ascending SBP categories (P < 0.001). As anticipated in an unadjusted observational context, the crude mortality rates for cardiovascular and cerebrovascular causes were disproportionately highest in the SBP ≥140 mmHg group (11.3 and 3.6%, respectively).
Table 1.
Baseline characteristics based on classification of systolic blood pressure for <130, 130–140, ≥140 mmHg
| Variables | Total | <130 mmHg | 130–139 mmHg | ≥140 mmHg | P |
|---|---|---|---|---|---|
| N | 2,198 | 1,040 | 434 | 724 | |
| Male | 1,225 (55.7) | 582 (56) | 241 (55.5) | 402 (55.5) | 0.979 |
| Age, years | 61.0 (49.0, 70.0) | 58.0 (45.8, 67.0) | 61.0 (51.0, 70.0) | 64.0 (53.0, 73.0) | <0.001 |
| Race, n (%) | <0.001 | ||||
| Mexican American | 446 (20.3) | 223 (21.4) | 68 (15.7) | 155 (21.4) | |
| Other Hispanic | 223 (10.1) | 102 (9.8) | 40 (9.2) | 81 (11.2) | |
| Non‐Hispanic White | 839 (38.2) | 427 (41.1)0 | 164 (37.8) | 248 (34.3) | |
| Non‐Hispanic Black | 513 (23.3) | 194 (18.7) | 128 (29.5) | 191 (26.4) | |
| Other Race | 177 (8.1) | 94 (9) | 34 (7.8) | 49 (6.8) | |
| Education, n (%) | 0.102 | ||||
| Less Than 9th Grade | 367 (16.7) | 158 (15.2) | 67 (15.4) | 142 (19.6) | |
| 9‐11th Grade | 365 (16.6) | 173 (16.6) | 73 (16.8) | 119 (16.4) | |
| High School Grad/GED or equivalent | 507 (23.1) | 234 (22.5) | 100 (23) | 174 (23.9) | |
| Some College or AA degree | 582 (26.5) | 287 (27.6) | 113 (26) | 183 (25.1) | |
| College Graduate or above | 374 (17.0) | 188 (18.1) | 81 (18.7) | 106 (14.5) | |
| Income, 10,000 dollars/year | 3.0 (1.8, 6.0) | 4.0 (2.0, 6.0) | 4.0 (1.8, 7.0) | 3.0 (1.8, 5.0) | <0.001 |
| Alcohol, g/day | 0.0 (0.0,0.0) | 0.0 (0.0,0.0) | 0.0 (0.0,0.0) | 0.0 (0.0,0.0) | 0.869 |
| BMI, kg/m2 | 30.8 (27.2, 35.7) | 30.8 (27.1, 35.9) | 31.2 (27.5, 36.1) | 30.7 (27.1, 35.3) | 0.305 |
| WC, cm | 107.0 (97.5, 117.7) | 106.5 (97.2, 117.7) | 107.9 (98.8, 119.3) | 106.9 (97.3, 116.3) | 0.064 |
| Total cholesterol, mg/dL | 184.8 (158.2, 216.2) | 186.0 (157.0, 216.9) | 181.0 (158.9, 208.8) | 186.0 (160.1, 218.3) | 0.146 |
| Triglyceride, mg/dL | 133.0 (93.9, 189.9) | 131.0 (93.9, 192.0) | 123.5 (87.2, 181.7) | 139.1 (98.2, 190.2) | 0.010 |
| HDL‐C, mg/dL | 47.2 (39.8, 56.8) | 46.0 (39.1, 56.1) | 47.2 (41.0, 56.8) | 47.2 (39.8, 59.2) | 0.094 |
| LDL‐C, mg/dL | 106.0 (82.0, 133.7) | 107.1 (81.0, 135.0) | 103.0 (83.0, 129.9) | 106.0 (83.0, 134.0) | 0.590 |
Data are presented as median (interquartile range, IQR) for continuous variables and as number (percentage) for categorical variables. P‐values were calculated using the Kruskal–Wallis test for continuous variables and the Chi‐square test for categorical variables. BMI, body mass index; HDL‐C, high‐density lipoprotein cholesterol; LDL‐C, low‐density lipoprotein cholesterol; TC, total cholesterol; TG, triglyceride; WC, waist circumference.
Table 2 delineates the baseline characteristics of the same cohort, alternatively stratified by DBP categories (<80, 80–89, ≥90 mmHg). Higher DBP levels were inversely associated with age; participants in the ≥90 mmHg group were significantly younger (median age 54.5 years) than those in the <80 mmHg group (median age 62.0 years; P < 0.001). Elevated DBP was also significantly associated with a higher BMI (P < 0.001), larger WC (P = 0.001), greater daily alcohol intake (P < 0.001), and a more adverse lipid profile, characterized by elevated levels of TC (P < 0.001) and LDL‐C (P < 0.001). Conversely, HDL‐C levels were lowest in the DBP ≥90 mmHg group (P = 0.001).
Table 2.
Baseline characteristics based on classification of diastolic blood pressure for <80, 80–90, ≥90 mmHg
| Variables | Total | <80 mmHg | 80–89 mmHg | ≥90 mmHg | P |
|---|---|---|---|---|---|
| N | 2,198 | 1,654 | 382 | 162 | |
| Male | 1,225 (55.7) | 900 (54.4) | 229 (59.9) | 96 (59.3) | 0.094 |
| Age, years | 61.0 (49.0, 70.0) | 62.0 (51.0, 71.0) | 56.0 (46.0, 66.0) | 54.5 (43.0, 65.8) | <0.001 |
| Race | <0.001 | ||||
| Mexican American | 446 (20.3) | 343 (20.7) | 72 (18.8) | 31 (19.1) | |
| Other Hispanic | 223 (10.1) | 166 (10) | 41 (10.7) | 16 (9.9) | |
| Non‐Hispanic White | 839 (38.2) | 675 (40.8) | 123 (32.2) | 41 (25.3) | |
| Non‐Hispanic Black | 513 (23.3) | 345 (20.9) | 108 (28.3) | 60 (37) | |
| Other Race | 177 (8.1) | 125 (7.6) | 38 (9.9) | 14 (8.6) | |
| Education | 0.430 | ||||
| Less Than 9th Grade | 367 (16.7) | 294 (17.8) | 49 (12.8) | 24 (14.8) | |
| 9‐11th Grade | 365 (16.6) | 270 (16.3) | 67 (17.5) | 28 (17.3) | |
| High School Grad/GED or equivalent | 507 (23.1) | 383 (23.1) | 82 (21.5) | 43 (26.5) | |
| Some College or AA degree | 582 (26.5) | 429 (25.9) | 114 (29.8) | 40 (24.7) | |
| College Graduate or above | 374 (17.0) | 278 (16.7) | 70 (18.3) | 27 (16.7) | |
| Income, 10,000 dollars/year | 3.0 (1.8, 6.0) | 3.0 (1.8, 6.0) | 4.0 (2.2, 6.8) | 4.0 (1.8, 6.0) | 0.048 |
| Alcohol, g/day | 0.0 (0.0, 0.0) | 0.0 (0.0, 0.0) | 0.0 (0.0, 0.0) | 0.0 (0.0, 0.0) | <0.001 |
| BMI, kg/m2 | 30.8 (27.2, 35.7) | 30.5 (26.9, 35.2) | 31.6 (27.6, 37.0) | 32.1 (28.3, 38.4) | <0.001 |
| WC, cm | 107.0 (97.5, 117.7) | 106.4 (97.2, 117.2) | 107.5 (98.3, 119.7) | 109.8 (101.8, 121.1) | 0.001 |
| Total cholesterol, mg/dL | 184.8 (158.2, 216.2) | 182.9 (156.1, 213.7) | 191.0 (165.9, 218.9) | 191.6 (168.4, 224.8) | <0.001 |
| Triglyceride, mg/dL | 133.0 (93.9, 189.9) | 131.0 (93.0, 188.0) | 140.5 (100.9, 195.0) | 142.9 (92.2, 188.9) | 0.068 |
| HDL‐C, mg/dL | 47.2 (39.8, 56.8) | 47.2 (39.8, 58.0) | 44.9 (39.1, 54.9) | 46.0 (38.0, 54.7) | 0.001 |
| LDL‐C, mg/dL | 106.0 (82.0, 133.7) | 103.0 (80.9, 131.0) | 112.6 (87.0, 140.0) | 114.8 (97.0, 144.5) | <0.001 |
Data are presented as median (interquartile range, IQR) for continuous variables and as number (percentage) for categorical variables. P‐values were calculated using the Kruskal–Wallis test for continuous variables and the Chi‐square test for categorical variables. BMI, body mass index; HDL‐C, high‐density lipoprotein cholesterol; LDL‐C, low‐density lipoprotein cholesterol; TC, total cholesterol; TG, triglyceride; WC, waist circumference.
Association between systolic blood pressure and mortality outcomes
To rigorously evaluate the independent association between SBP and mortality, we employed multivariable Cox proportional hazards regression models with comprehensive sequential adjustments for potential confounders, including DBP, sex, age, race, education, income, alcohol intake, BMI, waist circumference, and a complete lipid panel (TC, TG, HDL‐C, and LDL‐C) (Table 3).
Table 3.
Multivariate Cox regression on the relationship between systolic blood pressure (<130, 130–139, ≥140 mmHg) and all‐cause mortality, cardiovascular mortality, cerebrovascular mortality, and diabetes mortality
| Variable | Crude model | Adjusted model | ||
|---|---|---|---|---|
| HR | 95% CI | HR | 95% CI | |
| All‐cause mortality | ||||
| <130 mmHg | Ref. | Ref. | Ref. | Ref. |
| 130–139 mmHg | 1.08 | 0.78 ~ 1.51 | 0.97 | 0.69 ~ 1.37 |
| ≥140 mmHg | 1.90 | 1.47 ~ 2.44 | 1.66 | 1.28 ~ 2.16 |
| Cardiovascular mortality | ||||
| <130 mmHg | Ref. | Ref. | Ref. | Ref. |
| 130–139 mmHg | 1.02 | 0.67 ~ 1.54 | 1.69 | 1.23 ~ 2.31 |
| ≥140 mmHg | 0.95 | 0.62 ~ 1.45 | 1.53 | 1.10 ~ 2.13 |
| Cerebrovascular mortality | ||||
| <130 mmHg | Ref. | Ref. | Ref. | Ref. |
| 130–139 mmHg | 0.85 | 0.23 ~ 3.22 | 0.70 | 0.16 ~ 3.03 |
| ≥140 mmHg | 4.98 | 2.25 ~ 11.00 | 3.98 | 1.71 ~ 9.27 |
| Diabetes mortality | ||||
| <130 mmHg | Ref. | Ref. | Ref. | Ref. |
| 130–139 mmHg | 1.40 | 0.75 ~ 2.61 | 1.96 | 1.17 ~ 3.28 |
| ≥140 mmHg | 1.11 | 0.58 ~ 2.13 | 1.84 | 1.06 ~ 3.16 |
The crude model was unadjusted. The adjusted model was adjusted for DBP, sex, age, race, education, income, alcohol intake, BMI, WC, TC, TG, HDL‐C, and LDL‐C. BMI, body mass index; CI, confidence interval; DBP, diastolic blood pressure; HDL‐C, high‐density lipoprotein cholesterol; HR, hazard ratio; LDL‐C, low‐density lipoprotein cholesterol; SBP, systolic blood pressure; TC, total cholesterol; TG, triglycerides; WC, waist circumference.
According to the fully adjusted model, an SBP ≥140 mmHg was significantly and independently associated with a profoundly elevated risk across all evaluated mortality endpoints compared with the strictly controlled reference group (SBP <130 mmHg). Specifically, the adjusted hazard ratios (aHR) were as follows: all‐cause mortality (aHR = 1.66, 95% CI: 1.28–2.16, P < 0.001), cardiovascular mortality (aHR = 1.53, 95% CI: 1.10–2.13, P = 0.011), cerebrovascular mortality (aHR = 3.98, 95% CI: 1.71–9.27, P = 0.001), and diabetes‐related mortality (aHR = 1.84, 95% CI: 1.06–3.16, P = 0.029). Crucially, participants within the intermediate SBP range of 130–139 mmHg did not exhibit a statistically significant increase or decrease in mortality risk for any outcome compared to the <130 mmHg reference group, suggesting comparable long‐term survival outcomes between these two strata.
Association between diastolic blood pressure and mortality outcomes
The longitudinal analysis of DBP revealed a distinct and complex pattern indicative of a J‐curve relationship (Table 4). Following identical comprehensive multivariable adjustments (including SBP), participants with a DBP strictly between 80 and 89 mmHg demonstrated a significantly lower risk of mortality compared to the reference group with a DBP <80 mmHg. Specifically, the 80–89 mmHg category was associated with a 33% reduction in the relative risk of all‐cause mortality (aHR = 0.67, 95% CI: 0.46–0.98, P = 0.041) and a 43% reduction in cardiovascular mortality (aHR = 0.57, 95% CI: 0.35–0.93, P = 0.024). A DBP ≥90 mmHg was not significantly associated with reduced or increased mortality compared to the <80 mmHg group in the fully adjusted models. Furthermore, we observed no statistically significant independent correlation between DBP categories and either cerebrovascular mortality or diabetes‐specific mortality. Furthermore, when utilizing the intermediate range (130–139 mmHg for SBP and 80–89 mmHg for DBP) as the reference group, the analyses explicitly confirmed the lack of additional benefit for aggressive SBP lowering (Supplementary Tables S1 and 2).
Table 4.
Multivariate Cox regression on the relationship between diastolic blood pressure (<80, 80–89, ≥90 mmHg) and all‐cause mortality, cardiovascular mortality, cerebrovascular mortality, and diabetes mortality
| Variable | Crude model | Adjusted model | ||
|---|---|---|---|---|
| HR | 95% CI | HR | 95% CI | |
| All‐cause mortality | ||||
| <80 mmHg | Ref. | Ref. | Ref. | Ref. |
| 80–89 mmHg | 0.58 | 0.41 ~ 0.83 | 0.67 | 0.46 ~ 0.98 |
| ≥90 mmHg | 0.95 | 0.63 ~ 1.42 | 0.91 | 0.58 ~ 1.44 |
| Cardiovascular mortality | ||||
| <80 mmHg | Ref. | Ref. | Ref. | Ref. |
| 80–89 mmHg | 0.53 | 0.33 ~ 0.84 | 0.57 | 0.35 ~ 0.93 |
| ≥90 mmHg | 0.82 | 0.47 ~ 1.42 | 0.85 | 0.46 ~ 1.55 |
| Cerebrovascular mortality | ||||
| <80 mmHg | Ref. | Ref. | Ref. | Ref. |
| 80–89 mmHg | 0.28 | 0.07 ~ 1.18 | 0.51 | 0.12 ~ 2.25 |
| ≥90 mmHg | 1.88 | 0.78 ~ 4.54 | 2.54 | 0.88 ~ 7.28 |
| Diabetes mortality | ||||
| <80 mmHg | Ref. | Ref. | Ref. | Ref. |
| 80–89 mmHg | 0.83 | 0.45 ~ 1.56 | 0.99 | 0.51 ~ 1.92 |
| ≥90 mmHg | 0.96 | 0.41 ~ 2.23 | 0.84 | 0.33 ~ 2.17 |
The crude model was unadjusted. The adjusted model was adjusted for SBP, sex, age, race, education, income, alcohol intake, BMI, WC, TC, TG, HDL‐C, and LDL‐C. HR, hazard ratio; CI, confidence interval; DBP, diastolic blood pressure; SBP, systolic blood pressure; BMI, body mass index; WC, waist circumference; TC, total cholesterol; TG, triglycerides; HDL‐C, high‐density lipoprotein cholesterol; LDL‐C, low‐density lipoprotein cholesterol.
Kaplan–Meier survival estimates
The cumulative event rate curves generated via Kaplan–Meier survival analysis visually corroborated the findings derived from the Cox proportional hazards models. As illustrated in Figure 2, participants with an SBP ≥140 mmHg consistently displayed the steepest trajectory and highest cumulative incidence of all‐cause, cardiovascular, cerebrovascular, and diabetes‐related mortality over the follow‐up period (all log‐rank P < 0.05). Conversely, the survival trajectories for the <130 mmHg and 130–139 mmHg groups remained tightly clustered. When stratifying by DBP (Figure 3), the group with DBP levels <80 mmHg presented a significantly greater cumulative incidence of all‐cause and cardiovascular mortality compared to the 80–89 mmHg group.
Figure 2.
Kaplan–Meier survival curves for mortality endpoints stratified by systolic blood pressure (SBP). The panels display the cumulative event rates for (a) all‐cause mortality, (b) cardiovascular mortality, (c) cerebrovascular mortality, and (d) diabetes‐related mortality among participants with type 2 diabetes. The tables below each plot represent the number of participants at risk at the specified follow‐up times (in months). Statistical comparisons across the SBP categories (<130, 130–139, and ≥140 mmHg) were performed using the log‐rank test.
Figure 3.
Kaplan–Meier survival curves for mortality endpoints stratified by diastolic blood pressure (DBP). The panels display the cumulative event rates for (a) all‐cause mortality, (b) cardiovascular mortality, (c) cerebrovascular mortality, and (d) diabetes‐related mortality among participants with type 2 diabetes. The tables below each plot represent the number of participants at risk at the specified follow‐up times (in months). Statistical comparisons across the DBP categories (<80, 80–89, and ≥ 90 mmHg) were performed using the log‐rank test.
Stratified subgroup analysis
To assess the robustness and consistency of these epidemiological associations across diverse patient profiles, we conducted comprehensive stratified analyses by key demographic and clinical variables, including gender (male, female), age (<60 years, ≥60 years), and baseline BMI categories (<24 kg/m2, 24–28 kg/m2, >28 kg/m2). As depicted in the forest plots (Figure 4), the interaction tests yielded no significant results across any of the strata (P for interaction >0.05 for all interactions tested). This uniformity strongly indicates that the observed associations—specifically the heightened risk associated with SBP ≥140 mmHg and the comparatively favorable survival profile associated with a DBP of 80–89 mmHg—were highly consistent across these major demographic and clinical subgroups within the NHANES population.
Figure 4.
Subgroup analyses of the association between blood pressure categories and all‐cause mortality. The forest plots illustrate the hazard ratios (HRs) and 95% confidence intervals (CIs) for all‐cause mortality, stratified by sex, age (<60 vs ≥60 years), and body mass index (BMI) categories. (a) Subgroup analysis for SBP categories. (b) Subgroup analysis for DBP categories.
DISCUSSION
This retrospective cohort analysis elucidates the highly complex and nuanced epidemiological relationships between blood pressure thresholds and long‐term mortality outcomes in individuals with T2D, leveraging robust longitudinal data from the U.S. NHANES population. Our analyses consistently reveal two critical phenomena that challenge a “lower is always better” paradigm in diabetic hemodynamics. First, an SBP strictly maintained between 130 and 139 mmHg demonstrated mortality risks that were virtually indistinguishable from an SBP <130 mmHg. Second, we observed a distinct J‐curve relationship for DBP, wherein an intermediate range of 80–89 mmHg was associated with the lowest comparative risk of all‐cause and cardiovascular mortality compared with more aggressive reductions below 80 mmHg.
The SBP threshold: Observational evidence in T2D
Our most significant finding is the conspicuous absence of an additional mortality benefit for diabetic patients presenting with an SBP <130 mmHg compared to those situated in the 130–140 mmHg range. This observation adds a critical real‐world epidemiological context to the current guideline debate. While both the 2023 ESH and the 2025 ACC/AHA guidelines endorse a primary SBP target of <130 mmHg 8 , 10 , our observational data imply that the considerable pharmacological and clinical effort—along with the associated risks of polypharmacy—required to aggressively lower SBP from the 130–139 mmHg range to below 130 mmHg may not universally yield significant additional survival benefits for the broader diabetic population. This finding closely aligns with the landmark ACCORD‐BP and SPRINT subset analyses, which revealed no significant reduction in the composite rate of fatal and nonfatal major cardiovascular events when targeting an SBP <120 mmHg versus <140 mmHg in high‐risk diabetic patients 18 , 19 , 20 . Critically, however, our results also firmly establish an SBP ≥140 mmHg as an unequivocal risk threshold. An SBP ≥140 mmHg was profoundly associated with elevated risks across all mortality vectors, particularly cerebrovascular mortality. This underscores the absolute clinical imperative to ensure SBP is reliably controlled below 140 mmHg to mitigate accelerated endothelial dysfunction and macrovascular complications 21 , 22 .
The association between DBP and mortality
The observation that a DBP of 80–89 mmHg is associated with lower cardiovascular mortality than a DBP <80 mmHg highlights the intricate physiological paradox inherent in treating diabetic hypertension. The NHANES diabetic cohort is characterized by a high prevalence of obesity, which is inexorably linked to a greater underlying burden of subclinical atherosclerosis and CAD 23 , 24 . Because the left ventricular myocardium is perfused almost exclusively during diastole, excessive therapeutic reduction of DBP below 80 mmHg may critically compromise coronary perfusion pressure 17 , 25 , 26 . This risk of subendocardial ischemia is exponentially amplified in diabetic patients, whose coronary microvascular function and autoregulatory reserves are frequently already severely impaired by chronic hyperglycemia and lipotoxicity 21 , 27 .
Limitations
Despite its strengths, this study has several limitations. First, as an observational study, it is susceptible to residual confounding from unmeasured or imprecisely measured factors. We lacked complete data on lifestyle details and diabetes duration. Most notably, regarding antihypertensive treatment, we could only adjust for self‐reported medication use as a binary variable (yes/no). The NHANES dataset lacks granular, longitudinal information on the specific classes of antihypertensive drugs prescribed, their dosages, and the duration of therapy, which may independently impact survival and blood pressure trajectories. Second, findings from the U.S.‐based NHANES cohort may not be entirely generalizable to populations with distinct racial, socioeconomic, or metabolic profiles. Finally, blood pressure measurements were taken at a single baseline examination visit. While we utilized the average of three consecutive readings to minimize random error, single‐visit assessments cannot account for visit‐to‐visit variability and may not perfectly reflect long‐term “usual” blood pressure. This exposure misclassification inevitably introduces regression dilution bias, which typically attenuates observed epidemiological associations toward the null. Consequently, the true strength of the relationships we observed for both SBP and DBP might be underestimated. Specifically, regarding the J‐curve phenomenon observed for DBP, regression dilution bias may have flattened the risk curve, implying that the true mortality risk associated with excessively low DBP (<80 mmHg) could be even more pronounced than our current estimates suggest.
CONCLUSION
In summary, this retrospective cohort analysis identified associations between baseline blood pressure levels and long‐term survival outcomes in patients with type 2 diabetes. Baseline SBP <140 mmHg and DBP 80–89 mmHg were associated with more favorable mortality profiles. Given the observational design and reliance on single baseline measurements, these findings should be interpreted as associations rather than causal effects. They underscore the need for further studies, including randomized controlled trials, to inform individualized blood pressure management strategies that balance cardiovascular protection and organ perfusion.
DISCLOSURE
The authors declare that they have no conflicts of interest.
Approval of the research protocol: The study involving human participants (National Health and Nutrition Examination Survey, NHANES) was reviewed and approved by the Research Ethics Review Board of the National Center for Health Statistics.
Informed consent: The patients provided written informed consent to participate in this study. All participants signed informed consent forms. All methods were performed in accordance with relevant guidelines and regulations.
Registry and the registration no. of the study/trial: N/A.
Animal studies: N/A.
Supporting information
Table S1. Multivariate Cox regression on the relationship between systolic blood pressure (<130, 130–139, ≥140 mmHg) and all‐cause mortality, cardiovascular mortality, cerebrovascular mortality, and diabetes mortality.
Table S2. Multivariate Cox regression on the relationship between diastolic blood pressure (<80, 80–89, ≥90 mmHg) and all‐cause mortality, cardiovascular mortality, cerebrovascular mortality, and diabetes mortality.
ACKNOWLEDGMENTS
We thank all the participants and staff who participated in the NHANES.
Zhen‐Yu Liu and Zhan‐Ze Ma are co‐first authors.
Contributor Information
Zhen‐Yu Liu, Email: liuzhenyu200210@sjtu.edu.cn.
Fang‐Jun Xu, Email: xufangjundoctor@163.com.
DATA AVAILABILITY STATEMENT
This research was conducted through the NHANES from 1999 to 2018. The Centers for Disease Control and Prevention Institutional Review Board approved the NHANES protocols. The survey collects data from interviews and examinations through trained experts to create a representative sample via a complex multistage, random, stratified sample. URL: https://www.cdc.gov/nchs/nhanes/about‐data/index.html.
References
- 1. Ogurtsova K, da Rocha Fernandes JD, Huang Y, et al. IDF diabetes atlas: Global estimates for the prevalence of diabetes for 2015 and 2040. Diabetes Res Clin Pract 2017; 128: 40–50. [DOI] [PubMed] [Google Scholar]
- 2. Ferrannini E, Cushman WC. Diabetes and hypertension: The bad companions. Lancet 2012; 380: 601–610. [DOI] [PubMed] [Google Scholar]
- 3. Stamler J, Vaccaro O, Neaton JD, et al. Diabetes, other risk factors, and 12‐yr cardiovascular mortality for men screened in the multiple risk factor intervention trial. Diabetes Care 1993; 16: 434–444. [DOI] [PubMed] [Google Scholar]
- 4. Kannel WB, Wilson PW, Zhang TJ. The epidemiology of impaired glucose tolerance and hypertension. Am Heart J 1991; 121: 1268–1273. [DOI] [PubMed] [Google Scholar]
- 5. Adler AI, Stratton IM, Neil HA, et al. Association of systolic blood pressure with macrovascular and microvascular complications of type 2 diabetes (UKPDS 36): Prospective observational study. BMJ 2000; 321: 412–419. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Gross JL, de Azevedo MJ, Silveiro SP, et al. Diabetic nephropathy: Diagnosis, prevention, and treatment. Diabetes Care 2005; 28: 164–176. [DOI] [PubMed] [Google Scholar]
- 7. Thomas MC, Brownlee M, Susztak K, et al. Diabetic kidney disease. Nat Rev Dis Primers 2015; 1: 15018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Jones DW, Ferdinand KC, Taler SJ, et al. 2025 AHA/ACC/AANP/AAPA/ABC/ACCP/ACPM/AGS/AMA/ASPC/NMA/PCNA/SGIM guideline for the prevention, detection, evaluation and Management of High Blood Pressure in adults: A report of the American College of Cardiology/American Heart Association joint committee on clinical practice guidelines. Circulation 2025; 152: e114–e218. [DOI] [PubMed] [Google Scholar]
- 9. Bi Y, Li M, Liu Y, et al. Intensive blood‐pressure control in patients with type 2 diabetes. N Engl J Med 2024; 392: 1155–1167. [DOI] [PubMed] [Google Scholar]
- 10. Mancia G, Kreutz R, Brunström M, et al. 2023 ESH guidelines for the management of arterial hypertension the task force for the management of arterial hypertension of the European Society of Hypertension: Endorsed by the International Society of Hypertension (ISH) and the European renal association (ERA). J Hypertens 2023; 41: 1874–2071. [DOI] [PubMed] [Google Scholar]
- 11. Emdin CA, Rahimi K, Neal B, et al. Blood pressure lowering in type 2 diabetes: A systematic review and meta‐analysis. JAMA 2015; 313: 603–615. [DOI] [PubMed] [Google Scholar]
- 12. Beddhu S, Chertow GM, Greene T, et al. Effects of intensive systolic blood pressure lowering on cardiovascular events and mortality in patients with type 2 diabetes mellitus on standard glycemic control and in those without diabetes mellitus: Reconciling results from ACCORD BP and SPRINT. J Am Heart Assoc 2018; 7: e009326. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Buckley LF, Dixon DL, Wohlford GF, et al. Intensive versus standard blood pressure control in SPRINT‐eligible participants of ACCORD‐BP. Diabetes Care 2017; 40: 1733–1738. [DOI] [PubMed] [Google Scholar]
- 14. Duckworth W, Abraira C, Moritz T, et al. Glucose control and vascular complications in veterans with type 2 diabetes. N Engl J Med 2008; 360: 129–139. [DOI] [PubMed] [Google Scholar]
- 15. Liu Z, Zhai G. Cardiometabolic index and major depressive disorder: Stroke and diabetes as mediators. Prog Neuro‐Psychopharmacol Biol Psychiatry 2025; 138: 111340. [DOI] [PubMed] [Google Scholar]
- 16. Redon J, Mancia G, Sleight P, et al. Safety and efficacy of low blood pressures among patients with diabetes: Subgroup analyses from the ONTARGET (ONgoing Telmisartan alone and in combination with Ramipril global endpoint trial). J Am Coll Cardiol 2012; 59: 74–83. [DOI] [PubMed] [Google Scholar]
- 17. Vidal‐Petiot E, Ford I, Greenlaw N, et al. Cardiovascular event rates and mortality according to achieved systolic and diastolic blood pressure in patients with stable coronary artery disease: An international cohort study. Lancet 2016; 388: 2142–2152. [DOI] [PubMed] [Google Scholar]
- 18. Brunström M, Carlberg B. Association of blood pressure lowering with mortality and cardiovascular disease across blood pressure levels: A systematic review and meta‐analysis. JAMA Intern Med 2018; 178: 28–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Mariam A, Miller‐Atkins G, Pantalone KM, et al. A type 2 diabetes subtype responsive to ACCORD intensive glycemia treatment. Diabetes Care 2021; 44: 1410–1418. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Wright JT, Williamson JD, Whelton PK, et al. A randomized trial of intensive versus standard blood‐pressure control. N Engl J Med 2015; 373: 2103–2116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Horton WB, Barrett EJ. Microvascular dysfunction in diabetes mellitus and cardiometabolic disease. Endocr Rev 2021; 42: 29–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. UK prospective diabetes study group. BMJ 1998; 317: 703–713. [PMC free article] [PubMed] [Google Scholar]
- 23. Lowenstern A, Ng N, Takagi H, et al. Influence of obesity on coronary artery disease and clinical outcomes in the ADVANCE registry. Circ Cardiovasc Imaging 2023; 16: e014850. [DOI] [PubMed] [Google Scholar]
- 24. Zoungas S, Chalmers J, Neal B, et al. Follow‐up of blood‐pressure lowering and glucose control in type 2 diabetes. N Engl J Med 2014; 371: 1392–1406. [DOI] [PubMed] [Google Scholar]
- 25. Polese A, De Cesare N, Montorsi P, et al. Upward shift of the lower range of coronary flow autoregulation in hypertensive patients with hypertrophy of the left ventricle. Circulation 1991; 83: 845–853. [DOI] [PubMed] [Google Scholar]
- 26. Kjeldsen SE, Berge E, Bangalore S, et al. No evidence for a J‐shaped curve in treated hypertensive patients with increased cardiovascular risk: The VALUE trial. Blood Press 2015; 25: 83–92. [DOI] [PubMed] [Google Scholar]
- 27. Böhm M, Schumacher H, Teo KK, et al. Achieved diastolic blood pressure and pulse pressure at target systolic blood pressure (120‐140 mmHg) and cardiovascular outcomes in high‐risk patients: Results from ONTARGET and TRANSCEND trials. Eur Heart J 2018; 39: 3105–3114. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1. Multivariate Cox regression on the relationship between systolic blood pressure (<130, 130–139, ≥140 mmHg) and all‐cause mortality, cardiovascular mortality, cerebrovascular mortality, and diabetes mortality.
Table S2. Multivariate Cox regression on the relationship between diastolic blood pressure (<80, 80–89, ≥90 mmHg) and all‐cause mortality, cardiovascular mortality, cerebrovascular mortality, and diabetes mortality.
Data Availability Statement
This research was conducted through the NHANES from 1999 to 2018. The Centers for Disease Control and Prevention Institutional Review Board approved the NHANES protocols. The survey collects data from interviews and examinations through trained experts to create a representative sample via a complex multistage, random, stratified sample. URL: https://www.cdc.gov/nchs/nhanes/about‐data/index.html.