Association Between Antihypertensive Use and Hospitalized Pneumonia in Patients With Stroke: A Korean Nationwide Population-Based Cohort Study

Machine learning-based survival analysis approaches for predicting the risk of pneumonia post-stroke discharge

BACKGROUND

Post-stroke pneumonia (PSP) is common among stroke patients. PSP occurring after hospital discharge continues to increase the risk of poor functional outcomes and death among stroke survivors. Currently, there is no prediction model specifically designed to predict the occurrence of PSP beyond the acute stage of stroke. This study aimed to explore the use of machine learning (ML) methods in predicting the risk of PSP after hospital discharge.

METHODS

This study analyzed data from 5,754 hospitalized stroke patients. The dataset was randomly divided into a training set and a holdout test set, with a ratio of 80:20. Several clinical and laboratory variables were utilized as predictors and different ML algorithms were employed to model time-to-event data. The ML model's predictive performance was compared to existing risk-scoring systems. A model-agnostic method based on Shapley additive explanations was utilized to interpret the ML model.

RESULTS

The study found that 5.7% of the study patients experienced pneumonia within one year after discharge. Based on repeated 5-fold cross-validation on the training set, the random survival forest (RSF) model had the highest C-index among the various ML algorithms and traditional Cox regression analysis. The final RSF model achieved a C-index of 0.787 (95% confidence interval: 0.737-0.840) on the holdout test set, outperforming five existing risk-scoring systems. The top three important predictors were the Glasgow Coma Scale score, age, and length of hospital stay.

CONCLUSIONS

The RSF model demonstrated superior discriminative ability compared to other ML algorithms and traditional Cox regression analysis, suggesting a non-linear relationship between predictors and outcomes. The developed ML model can be integrated into the hospital information system to provide personalized risk assessments.

The Association between Statins Intake and Risk of Post Stroke Pneumonia: A Systematic Review and Meta-Analysis

AIM

This research aimed to examine the relationship between the intake of statins and the risk of post-stroke pneumonia in a systematic review and meta-analysis study.

METHODS

An extensive search of published articles on March 21st, 2023, was done in several databases, like Web of Science (ISI), PubMed, Cochrane Library, Embase, Scopus, and Google Scholar. The Newcastle Ottawa Scale (NOS) checklist was employed to evaluate the quality of observational studies. Statistical tests (Chi-square test and I2) and graphical techniques (Forest plot) were used to determine whether heterogeneity existed in the meta-analysis studies. Funnel plots and Begg and Egger's tests were used to assess the publication bias.

RESULTS

Seven studies (5 cohort and 2 case-control studies) were retrieved to examine the association between statins and post-stroke pneumonia. The sample size of the studies compiled in the meta- analysis was obtained to be 68,966 participants. Meta-analysis demonstrated that the overall odds of post-stroke pneumonia in the statin group was equal to 0.87 (95% CI: 0.67 - 1.13; p-value 0.458). Subgroup analysis indicated that the odds of post-stroke pneumonia in the statin group was equal to 0.93 (95% CI: 0.73-1.18; p-value = 0.558) in the cohort studies, and equal to 0.92 (95% CI: 0.37-2.26; p-value = 0.857) in the case-control studies. The examination of the association between the intake of statins and post-stroke pneumonia showed no evidence of publication bias (Begg's test, p-value = 0.368; Eggers test, p-value = 0.282).

CONCLUSION

In this study, no relationship has been observed between receiving statins and the risk of post-stroke pneumonia.

The Association Between Antihypertensive Drug Use and Hospitalization for Pneumonia in the General Population: A Case-Crossover Study Using the National Health Insurance Database of Korea

BACKGROUND

Previous studies have reported an association between pneumonia risk and the use of certain drugs. We investigated the relationship between antihypertensive drugs and pneumonia in the general population.

METHODS

This case-crossover study utilized the nationwide data of South Korea. We included participants who were hospitalized for pneumonia. A single case period was defined as 30 days before pneumonia onset, and two control periods were established (90-120 and 150-180 days before pneumonia onset). Further, we performed sensitivity and subgroup analyses (according to the presence of diabetes, documented disability, and whether participants were aged ≥ 70 years). We used conditional logistic regression models adjusted for covariates, such as angiotensin-converting-enzyme (ACE) inhibitors, angiotensin receptor blockers (ARBs), other antihypertensives, statins, antipsychotics, benzodiazepine, and the number of outpatient visits.

RESULTS

In total, 15,463 subjects were included in this study. ACE inhibitors (adjusted odds ratio [aOR], 0.660; 95% confidence interval [CI], 0.558-0.781), ARBs (aOR, 0.702; 95% CI, 0.640-0.770), and other antihypertensive drugs (aOR, 0.737; 95% CI, 0.665-0.816) were significantly associated with reduced pneumonia risk. Subgroup analyses according to the presence of diabetes mellitus, documented disability, and whether participants were aged ≥ 70 years consistently showed the association of antihypertensives with a reduced risk of hospitalization for pneumonia.

CONCLUSION

All antihypertensive drug types were related to a lower risk of hospitalization for pneumonia in the general population. Our results implied that frequent medical service usage and protective immunity were primarily related to a reduced risk of pneumonia in the general population of South Korea.