Abstract
Objectives : To identify variables associated with the development of pneumonia in patients with ischemic stroke and to assess the utility of both the clinical swallowing examination and the videofluoroscopic modified barium swallow (VMBS) to identify stroke patients at risk for aspiration and subsequent pneumonia and to measure the effect of pneumonia on hospital outcomes.
Design/Methods: Bedside clinical evaluation and VMBS were performed on 102 patients admitted with acute ischemic stroke who were referred for swallowing eval uation because of clinical suspicion of dysphagia. The clinical features, stroke loca tion and severity, pneumonia, length of stay (LOS), mortality, and costs were evalu ated retrospectively.
Results: Pneumonia occurred in 21 of 102 patients (20.6%) and was more fre quent in those with aspiration on VMBS (p < 0.01). Those with pneumonia had longer median LOS (29 days vs. 10 days), higher total costs ($27,764 vs. $9,753, p < 0.0001), higher mortality (24 percent vs. 3.7 percent, p < 0.01), and worse 24-hour stroke severity scores (p = 0.001).
The VMBS correlated better with developing pneumonia than did the clinical swallowing assessment alone, but the combined use of both tests yielded the highest sensitivity (0.86) and a negative predictive value (0.91) for pneumonia.
Conclusions: Pneumonia complicating ischemic stroke significantly correlates with increased mortality, increased LOS, higher costs, and 24-hour stroke severity scores. Pneumonia was more frequent in patients who aspirated on VMBS and in pa tients who had an impaired level of consciousness. The combined use of clinical swal lowing evaluation and VMBS has high negative predictive value in assessing risk for pneumonia.