Pulse pressure and stroke risk: development and validation of a new stroke risk model

Exploring the relationship between admission pulse pressure and clinical features in patients with spontaneous supratentorial intracerebral hemorrhage

Elevated pulse pressure is commonly observed in cardiovascular diseases and serves as an independent risk factor and predictor of cardiac mortality. However, the role of pulse pressure in patients with spontaneous intracerebral hemorrhage (ICH) remains uncertain. This study aimed to investigate the association between admission pulse pressure and clinical characteristics, including in-hospital outcomes, in ICH patients. We retrospectively analyzed the data of 292 ICH patients, categorizing them into two groups based on admission wide pulse pressure: > 100 mmHg (n = 60) and ≤ 100 mmHg (n = 232). Clinical characteristics and in-hospital outcomes were compared between the groups, and multivariate logistic regression was performed to identify independent factors. Patients with wide pulse pressure were older, had lower Glasgow Coma Scale, larger intraparenchymal hematomas, more pronounced midline shifts, and higher rates of intraventricular hematoma extension and hydrocephalus. These patients also experienced higher frequencies of craniotomy or craniectomy and longer hospital stays. Multivariate logistic regression revealed that pulse pressure > 100 mmHg was significantly associated with increased in-hospital mortality (odds ratio 4.31, 95% confidence interval 1.12-16.62, p = 0.03), but not with a modified Rankin Scale score of 4-6. In conclusion, our investigation demonstrates a significant relationship between admission pulse pressure and severe clinical characteristics in ICH patients. Importantly, a wider pulse pressure is linked to heightened in-hospital mortality. These results underscore the necessity for customized strategies to predict patient outcomes in this population. Further research is essential to explore potential therapeutic interventions targeting pulse pressure to improve clinical outcomes for ICH patients.

A Single-Center Experience of Correlation of Pulse Pressure to Mortality of Stroke Hemorrhage Patients in Indonesia

Introduction

The relationship between pulse pressure and mortality in acute stroke hemorrhage patients is a subject of debate. To investigate this relationship in the Indonesian context, a study was conducted due to the increasing prevalence of stroke in the country.

Methods

The study sample consisted of 111 patients with acute stroke hemorrhage admitted to the hospital between January 1, 2016, and December 31, 2019. Patients with sepsis, cancer, or other hematology disorders were excluded, as were those who were lost to follow-up. Statistical analysis was performed using SPSS 22, and correlations were evaluated between various patient characteristics and laboratory values.

Results

It was revealed that patients with a wider pulse pressure were more likely to die (adjusted odds ratio = 3,070) than those with a normal or constricted pulse pressure.

Conclusion

Pulse pressure had an impact on the mortality of patients with acute hemorrhagic stroke.

Effect of Grading Rehabilitation Nursing Mode on Limb Function, Speech Rehabilitation, and Quality of Life of Stroke Patients

Objective. The aim of this study is to investigate the influence of grading rehabilitation nursing mode on limb function, speech function, and QOL of stroke patients. Methods. From January 2018 to April 2019, the stroke patients who received treatment in our hospital were selected as the study participants. Based on the random number table, they were assigned to CG (n = 60) and OG (n = 60). The routine rehabilitation nursing mode was used in the CG, and the grading rehabilitation nursing mode was used in the OG. The limb function, speech function, QOL, and nursing satisfaction were evaluated in both the groups, and the survival curve was analyzed after 12 months of follow-up. Results. The motor function of upper and lower limbs in OG was significantly higher than that in CG, and the total effective rate of speech function recovery in OG was 95.00%, which was obviously higher than 81.67% in CG ( $P<0.05$ ); the total QOL score in OG was (80.72 ± 7.15), which was significantly higher than (69.53 ± 6.42) in CG. The nursing satisfaction of the OG was higher ( $P<0.05$ ). The Kaplan–Meier curve analysis revealed that the difference of 12-month survival rate between CG and OG was statistically significant (χ2 = 4.710, $P=0.030$ ). Conclusion. The application of grading rehabilitation nursing mode in stroke patients can effectively facilitate the recovery of extremity function and speech function, ameliorate the QOL and nursing satisfaction of patients, reduce the death and disability of patients, and prolong the survival time of patients.

Identification of important factors in an inpatient fall risk prediction model to improve the quality of care using EHR and electronic administrative data: A machine-learning approach

BACKGROUND

Inpatient falls, many resulting in injury or death, are a serious problem in hospital settings. Existing falls risk assessment tools, such as the Morse Fall Scale, give a risk score based on a set of factors, but don't necessarily signal which factors are most important for predicting falls. Artificial intelligence (AI) methods provide an opportunity to improve predictive performance while also identifying the most important risk factors associated with hospital-acquired falls. We can glean insight into these risk factors by applying classification tree, bagging, random forest, and adaptive boosting methods applied to Electronic Health Record (EHR) data.

OBJECTIVE

The purpose of this study was to use tree-based machine learning methods to determine the most important predictors of inpatient falls, while also validating each via cross-validation.

MATERIALS AND METHODS

A case-control study was designed using EHR and electronic administrative data collected between January 1, 2013 to October 31, 2013 in 14 medical surgical units. The data contained 38 predictor variables which comprised of patient characteristics, admission information, assessment information, clinical data, and organizational characteristics. Classification tree, bagging, random forest, and adaptive boosting methods were used to identify the most important factors of inpatient fall-risk through variable importance measures. Sensitivity, specificity, and area under the ROC curve were computed via ten-fold cross validation and compared via pairwise t-tests. These methods were also compared to a univariate logistic regression of the Morse Fall Scale total score.

RESULTS

In terms of AUROC, bagging (0.89), random forest (0.90), and boosting (0.89) all outperformed the Morse Fall Scale (0.86) and the classification tree (0.85), but no differences were measured between bagging, random forest, and adaptive boosting, at a p-value of 0.05. History of Falls, Age, Morse Fall Scale total score, quality of gait, unit type, mental status, and number of high fall risk increasing drugs (FRIDs) were considered the most important features for predicting inpatient fall risk.

CONCLUSIONS

Machine learning methods have the potential to identify the most relevant and novel factors for the detection of hospitalized patients at risk of falling, which would improve the quality of patient care, and to more fully support healthcare provider and organizational leadership decision-making. Nurses would be able to enhance their judgement to caring for patients at risk for falls. Our study may also serve as a reference for the development of AI-based prediction models of other iatrogenic conditions. To our knowledge, this is the first study to report the importance of patient, clinical, and organizational features based on the use of AI approaches.

Prediction models for cardiovascular disease risk in the hypertensive population: a systematic review

OBJECTIVE

The aim of this study was to identify, describe, and evaluate the available cardiovascular disease risk prediction models developed or validated in the hypertensive population.

METHODS

MEDLINE and the Web of Science were searched from database inception to March 2019, and all reference lists of included articles were reviewed.

RESULTS

A total of 4766 references were screened, of which 18 articles were included in the review, presenting 17 prediction models specifically developed for hypertensive populations and 25 external validations. Among the 17 prediction models, most were constructed based on randomized trials in Europe or North America to predict the risk of fatal or nonfatal cardiovascular events. The most common predictors were classic cardiovascular risk factors such as age, diabetes, sex, smoking, and SBP. Of the 17 models, only one model was externally validated. Among the 25 external validations, C-statistics ranged from 0.58 to 0.83, 0.56 to 0.75, and 0.64 to 0.78 for models developed in the hypertensive population, the general population and other specific populations, respectively. Most of the development studies and validation studies had an overall high risk of bias according to PROBAST.

CONCLUSION

There are a certain number of cardiovascular risk prediction models in patients with hypertension. The risk of bias assessment showed several shortcomings in the methodological quality and reporting in both the development and validation studies. Most models developed in the hypertensive population have not been externally validated. Compared with models developed for the general population and other specific populations, models developed for the hypertensive population do not display a better performance when validated among patients with hypertension. Research is needed to validate and improve the existing cardiovascular disease risk prediction models in hypertensive populations rather than developing completely new models.

Low Diastolic Blood Pressure is Not Related to Risk of First Episode of Stroke in a High-Risk Population: A Secondary Analysis of SPRINT

Background

Hypertension is the most prevalent and leading risk factor for stroke. SPRINT (The Systolic Blood Pressure Intervention Trial) assessed the effects on cardiovascular event risk of intensive compared with standard systolic blood pressure reduction. In this secondary analysis of SPRINT data, we investigated how low on‐treatment diastolic blood pressure (DBP) influenced risk for stroke events.

Methods and Results

For this analysis, we used SPRINT_POP (Primary Outcome Paper) Research Materials from the National Heart, Lung and Blood Institute (NHLBI) Biologic Specimen and Data Repository Information Coordinating Center. Data for 8944 SPRINT participants were analyzed from the period after target blood pressure was achieved until the end of the trial. Overall, there were 110 stroke events, including 49 from the intensive‐treatment arm and 61 in the standard‐treatment group. In participants with DBP <70 mm Hg, stroke risk was higher than with DBP ≥70 mm Hg (hazard ratio, 1.467; 95% CI 1.009–2.133; P=0.0445). Univariable Cox proportional hazard risk analysis showed that in the whole group, age and cardiovascular and chronic renal diseases were stroke risk factors. These risk factors were related to lower DBP and higher pulse pressure, however, not to study arm. Multivariable Cox proportional hazard analysis revealed that only age, history of cardiovascular disease, current smoking status and on‐treatment systolic blood pressure were significantly related to stroke risk.

Conclusions

Low on‐treatment DBP is not related to the risk for the first stroke, in contrast to older age, the history of cardiovascular disease, current smoking status, and on‐treatment systolic blood pressure.

Clinical Trial Registration

URL: https://www.clinicaltrials.gov. Unique identifier: NCT01206062.

Associations between height and blood pressure in the United States population

The mechanisms linking short stature with an increase in cardiovascular and cerebrovascular disease risk remain elusive. This study tested the hypothesis that significant associations are present between height and blood pressure in a representative sample of the US adult population.Participants were 12,988 men and women from a multiethnic sample (age ≥ 18 years) evaluated in the 1999 to 2006 National Health and Nutrition Examination Survey who were not taking antihypertensive medications and who had complete height, weight, % body fat, and systolic and diastolic arterial blood pressure (SBP and DBP) measurements; mean arterial blood pressure and pulse pressure (MBP and PP) were calculated. Multiple regression models for men and women were developed with each blood pressure as dependent variable and height, age, race/ethnicity, body mass index, % body fat, socioeconomic status, activity level, and smoking history as potential independent variables.Greater height was associated with significantly lower SBP and PP, and higher DBP (all P < .001) in combined race/ethnic-sex group models beginning in the 4th decade. Predicted blood pressure differences between people who are short and tall increased thereafter with greater age except for MBP. Socioeconomic status, activity level, and smoking history did not consistently contribute to blood pressure prediction models.Height-associated blood pressure effects were present in US adults who appeared in the 4th decade and increased in magnitude with greater age thereafter. These observations, in the largest and most diverse population sample evaluated to date, provide support for postulated mechanisms linking adult stature with cardiovascular and cerebrovascular disease risk.

Elevated Pulse Pressure Levels Are Associated With Increased In-Hospital Mortality in Acute Spontaneous Intracerebral Hemorrhage

OBJECTIVES

Clinical outcome after intracerebral hemorrhage (ICH) remains poor. Definitive phase-3 trials in ICH have failed to demonstrate improved outcomes with intensive systolic blood pressure (SBP) lowering. We sought to determine whether other BP parameters-diastolic BP (DBP), pulse pressure (PP), and mean arterial pressure (MAP)-showed an association with clinical outcome in ICH.

METHODS

We retrospectively analyzed a prospective cohort of 672 patients with spontaneous ICH and documented demographic characteristics, stroke severity, and neuroimaging parameters. Consecutive hourly BP recordings allowed for computation of SBP, DBP, PP, and MAP. Threshold BP values that transitioned patients from survival to death were determined from ROC curves. Using in-hospital mortality as outcome, BP parameters were evaluated with multivariable logistic regression analysis.

RESULTS

Patients who died during hospitalization had higher mean PP compared to survivors (68.5 ± 16.4 mm Hg vs. 65.4 ± 12.4 mm Hg; P = 0.032). The following admission variables were associated with significantly higher in-hospital mortality (P < 0.001): poorer admission clinical condition, intraventricular hemorrhage, and increased admission normalized hematoma volume. ROC analysis showed that mean PP dichotomized at 72.17 mm Hg, provided a transition point that maximized sensitivity and specific for mortality. The association of this increased dichotomized PP with higher in-hospital mortality was maintained in multivariable logistic regression analysis (odds ratio, 3.0; 95% confidence interval, 1.7-5.3; P < 0.001) adjusting for potential confounders.

CONCLUSION

Widened PP may be an independent predictor for higher mortality in ICH. This association requires further study.

Opportunities and challenges in developing risk prediction models with electronic health records data: a systematic review

OBJECTIVE

Electronic health records (EHRs) are an increasingly common data source for clinical risk prediction, presenting both unique analytic opportunities and challenges. We sought to evaluate the current state of EHR based risk prediction modeling through a systematic review of clinical prediction studies using EHR data.

METHODS

We searched PubMed for articles that reported on the use of an EHR to develop a risk prediction model from 2009 to 2014. Articles were extracted by two reviewers, and we abstracted information on study design, use of EHR data, model building, and performance from each publication and supplementary documentation.

RESULTS

We identified 107 articles from 15 different countries. Studies were generally very large (median sample size = 26 100) and utilized a diverse array of predictors. Most used validation techniques (n = 94 of 107) and reported model coefficients for reproducibility (n = 83). However, studies did not fully leverage the breadth of EHR data, as they uncommonly used longitudinal information (n = 37) and employed relatively few predictor variables (median = 27 variables). Less than half of the studies were multicenter (n = 50) and only 26 performed validation across sites. Many studies did not fully address biases of EHR data such as missing data or loss to follow-up. Average c-statistics for different outcomes were: mortality (0.84), clinical prediction (0.83), hospitalization (0.71), and service utilization (0.71).

CONCLUSIONS

EHR data present both opportunities and challenges for clinical risk prediction. There is room for improvement in designing such studies.