Association of Systolic Blood Pressure after Discharge and the Risk of Clinical Outcomes in Ischemic Stroke Patients with Diabetes: a Cohort Study
Pinni Yang,Zhengbao Zhu,Shuyao Wang,Mengyao Shi,Yanbo Peng,Chongke Zhong,Aili Wang,Tan Xu,Hao Peng,Tian Xu,Xiaowei Zheng,Jing Chen,Yonghong Zhang,Jiang He
DOI: https://doi.org/10.1097/cm9.0000000000002819
IF: 6.133
2023-01-01
Chinese Medical Journal
Abstract:To the Editor: Hypertension is an important risk factor for the incidence and prognosis of diabetes and ischemic stroke. Both diabetes and ischemic stroke are common disorders that often occur together, and up to approximately 34% of acute ischemic stroke patients also experience diabetes.[1] Previous studies have shown that stroke patients with hyperglycemia or diabetes exhibited increased risks of recurrent stroke, poor neurological outcomes, and mortality. However, few studies have explored the relationship between systolic blood pressure (SBP) levels or SBP trajectories and clinical outcomes in patients with both ischemic stroke and diabetes. Herein, this study aimed to investigate the associations of SBP at 14 days or hospital discharge, SBP at 12 months after stroke onset, and the SBP trajectory after discharge with the risks of adverse clinical outcomes over a period from 12 to 24 months among ischemic stroke patients with diabetes from the China Antihypertensive Trial in Acute Ischemic Stroke (CATIS)[2]. This study was based on the CATIS, and complete information regarding the CATIS design, procedures, and main results have been described previously. The patient selection process of the present study was shown in Supplementary Figure 1, https://links.lww.com/CM9/B680. Ischemic stroke patients complicated with diabetes were identified according to the following criteria: Fasting blood glucose concentration ≥7.0 mmol/L, receiving hypoglycemic treatment during hospitalization or self-reported history of physician-diagnosed diabetes.[3] The CATIS protocol was approved by the institutional review boards at Soochow University in China (No. Soochow Ethics 2007-01) and Tulane University in the United States as well as ethical committees at each participating hospital. Written consent was provided by all participants or their immediate family members. Baseline information was collected at the time of enrollment. Stroke severity was assessed by a trained neurologist at baseline using the National Institutes of Health Stroke Scale (NIHSS). BP were measured at baseline, 14 days or hospital discharge, three months, and 12 months after stroke onset by trained nurses according to procedures recommended by the American Heart Association. The average value of the three measurements at each time point was obtained as the clinic BP and used in the analyses. Dyslipidemia was defined according to the Chinese guidelines on the prevention and treatment of dyslipidemia. The estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration creatinine equation with an adjusted coefficient of 1.1 for the Chinese population. To take advantage of repeated BP assessments, the SAS PROC TRAJ procedure (SAS Institute, Cary, NC, USA) was used for latent mixture modeling to identify trajectories that share similar long-term SBP patterns from 14 days or hospital discharge to 12 months after ischemic stroke with the greatest posterior probability.[4] We initiated a model with one trajectory and then fitted models up to the optimal number of trajectories by comparing the Bayesian information criterion. The model with four trajectories was identified as exhibiting the best balance between data fit and complexity (range of average likelihood across trajectories: 0.78–0.90). We assigned a descriptive label for each trajectory group based on the visual patterns of SBP changes over time: low-stable, high-decreasing, moderate-increasing, and persistent-high [Figure 1].Figure 1: Trajectories of SBP among ischemic stroke patients with diabetes in the CATIS study. Three BP measurements were obtained at 14 days or hospital discharge, 3 months and 12 months after stroke onset. The mean of three SBP measurements at each of the time points was used to fit trajectories. BP: Blood pressure; CATIS: China Antihypertensive Trial in Acute Ischemic Stroke; SBP: Systolic blood pressure.The primary outcome was a combination of death and vascular events (i.e., vascular deaths, recurrent nonfatal stroke, nonfatal myocardial infarction, hospitalized and treated angina, hospitalized and treated congestive heart failure, and hospitalized and treated peripheral arterial disease) from 12 to 24 months after stroke onset. Secondary outcomes included death, vascular events, and major disability (modified Rankin Scale [mRS] score ≥3) from 12 to 24 months after stroke onset. Information on all vascular events was obtained from hospital data, and death certificates were obtained for deceased patients. Participants were categorized into three groups according to SBP at 14 days or hospital discharge or SBP at 12 months: ≥160 mmHg, 140–160 mmHg (inclusive of 140 mmHg), and <140 mmHg. Odds ratios (ORs), hazard ratios (HRs), and 95% confidence intervals (CIs) were calculated using multivariate logistic regression and Cox proportional hazards models as appropriate for clinical outcomes according to SBP groups and SBP trajectories. We adjusted for age and sex in the initial models. Important covariates for the prognosis of ischemic stroke, including time from onset to hospitalization, baseline SBP, high-sensitivity C-reactive protein, eGFR, and NIHSS score at baseline, dyslipidemia, ischemic stroke subtypes, history of hypertension, history of coronary heart disease, family history of stroke, and receiving immediate BP reduction and occupation, were selected based on our prior knowledge and further included in the multivariable adjusted model. We conducted several sensitivity analyses by further adjusting the use of lipid-lowering drugs before admission or anticoagulation therapy during hospitalization or antiplatelet therapy during hospitalization or excluding patients with eGFR <30 mL∙min-1∙1.73 m-2 (chronic kidney disease stage 4–5) to estimate the association of SBP trajectories with the prognosis in patients with diabetes and ischemic stroke. Two-sided P <0.05 indicated statistical significance. All statistical analyses were performed with SAS version 9.4 (SAS Institute). The baseline characteristics of patients included in the analyses for SBP at 14 days or hospital discharge and at 12 months according to the three SBP groups are shown in Supplementary Tables 1 and 2, https://links.lww.com/CM9/B680. The multivariate-adjusted HRs (95% CIs) of participants with SBP <140 mmHg were 0.54 (0.35–0.84) for the primary outcome and 0.53 (0.30–0.95) for death compared with participants with SBP ≥160 mm-Hg at 14 days or hospital discharge. The multivariate-adjusted HRs associated with SBP <140 mmHg at 12 months were 0.39 (95% CI, 0.20–0.78) for the primary outcome, 0.35 (95% CI, 0.17–0.73) for vascular events, and 0.39 (95% CI, 0.20–0.73) for major disability [Supplementary Table 3, https://links.lww.com/CM9/B680]. We identified four distinct SBP trajectories from 14 days or hospital discharge to 12 months after ischemic stroke among 1156 participants [Figure 1]. Baseline characteristics, such as age, baseline SBP, eGFR, baseline NIHSS score, and receiving immediate BP reduction, significantly differed across these four SBP trajectories [Supplementary Table 4, https://links.lww.com/CM9/B680]. From 12 to 24 months after stroke onset, 68 (5.9%) participants experienced the primary outcome, 52 (4.5%) participants developed vascular events, and 29 (2.5%) participants died. After adjusting for age, sex, baseline SBP, NIHSS score at baseline, and other important prognostic factors, the HRs of the primary outcome were 0.27 (95% CI, 0.09–0.81) for the moderate-increasing pattern, 0.26 (95% CI, 0.07–0.96) for the high-decreasing pattern, and 0.15 (95% CI, 0.05–0.45) for the low-stable pattern. The multivariable-adjusted HRs of death were 0.16 (95% CI, 0.03–0.85) for the moderate-increasing pattern, 0.14 (95% CI, 0.02–0.96) for the high-decreasing pattern, and 0.06 (95% CI, 0.01–0.34) for the low-stable pattern compared with the persistent-high pattern [Supplementary Table 5, https://links.lww.com/CM9/B680]. In the sensitivity analyses, the association between SBP trajectories and primary outcome remained significant [Supplementary Table 6, https://links.lww.com/CM9/B680]. In this prospective study, we observed that patients with both ischemic stroke and diabetes whose SBP achieved<140 mmHg at 14 days or hospital discharge or at 12 months had a significantly lower risk of primary outcome in the period from 12 to 24 months after stroke onset. In addition, the patients with a low-stable SBP trajectory had the lowest risk of unfavorable outcomes. Moreover, the patients with the high-decreasing trajectory had higher SBP levels at discharge that decreased to lower SBP levels in the late follow-up period than those with the moderate-increasing pattern but a lower risk of the unfavorable outcomes in the period from 12 to 24 months after stroke onset. These findings suggested that controlling SBP after discharge might be beneficial in improving the long-term prognosis among ischemic stroke patients with diabetes. Hypertensive patients with diabetes but not stroke with a target SBP of 120 mmHg exhibited a higher risk of cardiovascular events than those with a goal of 140 mmHg in the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study. Conversely, the Secondary Prevention of Small Subcortical Strokes (SPS3) trial focused on lacunar stroke patients with diabetes and suggested that SBP lower than 130 mmHg would be particularly beneficial for the prevention of recurrent stroke compared with the target of SBP of 130–149 mmHg. In the Systolic Blood Pressure Intervention Trial (SPRINT) study, decreased risks of major cardiovascular events and all-cause mortality were found in the individuals randomized to intensive lowering BP treatment (SBP target <120 mmHg) compared with standard lowering BP treatment (SBP target <140 mmHg).[5] Of note, individuals with diabetes or stroke were excluded from the SPRINT. Thus, whether the findings of SPRINT could be directly transferable to patients with stroke or diabetes remained unknown. Most recently, Zheng et al[6] reported that patients with a high-decreasing (SBP of 164 mmHg after discharge that decreased to 140 mmHg at 12 months) or low-stable SBP trajectory (SBP consistently <140 mmHg during the follow-up) from discharge to one year after stroke had a lower risk of unfavorable clinical outcomes among ischemic stroke patients, but the specific association of SBP trajectories with outcomes in ischemic stroke patients with diabetes was not observed in their analyses. Moreover, no guideline recommendations for BP control have been made for stroke patients with comorbid diabetes. Based on the participants with diabetes in CATIS, we conducted a prospective multicenter study to examine the association between post-discharge SBP control and unfavorable clinical outcomes in patients with both ischemic stroke and diabetes. Standardized protocols and strict quality control procedures were used in baseline data collection, BP measurement, and outcome assessment during follow-up. Furthermore, comprehensive information about potential confounders was collected and controlled in the multivariate models. Therefore, our study would provide more credible evidence for the association between post-discharge SBP control and unfavorable clinical outcomes among ischemic stroke patients with diabetes. Our study described the pattern of association of long-term SBP change with unfavorable clinical outcomes after discharge among ischemic stroke patients with diabetes. The patients with SBP <140 mmHg at discharge and 12 months after stroke as well as the patients with low-stable discharge SBP trajectories with consistently SBP of 135 mmHg had the lowest risk of primary outcome. Sensitivity analyses further confirmed these findings. Therefore, our study suggested that stabilization of SBP between 130mm Hg and 140 mmHg was safe and might be most beneficial for ischemic stroke patients with diabetes. However, whether SBP could be reduced to less than 120 mmHg should be investigated in further clinical trials. The mechanism by which post-discharge BP control might improve clinical outcomes in ischemic stroke patients with diabetes is currently unclear. It is possible that BP control after discharge may be particularly associated with less edema and a lower risk of bleeding, which may be beneficial for outcomes. Our study had several limitations. First, as an observational study, we were unable to eliminate residual confounding factors completely, although several important confounders were adjusted in multivariate analysis. Second, only 19 patients were assigned a persistent-high trajectory, and the sample size might be relatively small to obtain a sufficient incidence rate of clinical outcomes in the persistent-high trajectory group. However, the four SBP trajectories identified by the principle of latent mixture modeling were reasonable. Third, some patients could experience stress hyperglycemia during the acute phase of stroke but not diabetes. However, it was reported that patients with stress hyperglycemia faced worse consequences, and approximately 60% of patients with admission hyperglycemia were confirmed to have diabetes at one year.[7] In conclusion, we found that better post-discharge SBP control was associated with a lower risk of the composite outcome of death and vascular events among ischemic stroke patients with diabetes. Patients with a low-stable SBP trajectory pattern had the lowest risk of poor prognosis. Our results provide further motivation for randomized clinical trials that explicitly establish the optimal target and range of post-discharge SBP for patients with both ischemic stroke and diabetes. Funding This study was supported by a grant from the National Natural Science Foundation of China (No. 82020108028). Conflicts of interest None.