An Inpatient Scoring System To Prevent Hospital Acquired Pneumonia on Stroke Service (4705)

Objective: To develop a scoring metrics of various risk factors to identify stroke patients at risk for developing pneumonia. Background: Aspiration pneumonia is the result of inhalation of oropharyngeal contents into the lower airways. It is the number one cause of infection in patients with stroke. It is also associated with an increased mortality, morbidity and poor functional outcomes. It is estimated that 6%–22% of stroke patients suffer from pneumonia. The purpose of the study is to determine different risk factors and conditions that would help identify high-risk patients for aspiration pneumonia using a scoring system. Design/Methods: Electronic medical record (EMR) data from January 2016 to June 2016 was extracted and analyzed. ICD-10 codes were used to search the diagnosis of Ischemic Stroke (IS) and pneumonia. Aspiration pneumonia was considered if pneumonia was reported within 30 days of stroke. Co-morbid conditions and other variables included age, asthma, heart disease, heart failure, atrial fibrillation, CKD, diabetes, hypertension, COPD, dysphagia, dementia and delirium. Statistical modeling using two-tailed chi-square test was employed to find the best-fitted model for the scoring system. Results: A total of 376 patients were identified with acute ischemic stroke. Among these, 33 patients (9.12%) were found to have post stroke aspiration pneumonia. The mean age was 67 years with majority of the patients being females 21 (58%). The multivariable model for the predictor variables included; age > 80 years, presence of chronic obstructive pulmonary disease (COPD), dysphagia, asthma, heart failure, dementia and delirium to have the highest sensitivity (93.9%), specificity (57%) and negative predictive value (98.9%). Conclusions: Presence of any one of the following co-morbidities can select patients at high risk of developing pneumonia on stroke service; COPD, Heart failure, Asthma, Age >80, Dysphagia, Delirium, and Dementia. This method has high sensitivity and NPV. This is a preliminary report from a larger prospective study. Disclosure: Dr. Farooqui has nothing to disclose. Dr. Farasatpour has nothing to disclose. Dr. Suriya has nothing to disclose. Dr. Qaryouti has nothing to disclose. Dr. Zafar has nothing to disclose.