Prognostic implications of swallowing ability in elderly patients after initial recovery from stroke.

BACKGROUND It remains unclear how swallowing assessment can help clinicians to predict the risk for pneumonia in elderly persons after ischemic stroke. A prospective case-control study was conducted to evaluate the prognostic utility of swallowing ability assessments. METHODS Participants were 136 elderly persons who had an acute ischemic stroke 3-12 months previously. They were separated into four groups based on their history of repeated episodes of pneumonia in combination with swallowing ability: Group 1 had neither repeated pneumonia nor swallowing abnormality (n = 69); group 2 had repeated pneumonia but no swallowing abnormality (n = 0); group 3 had swallowing abnormality but no repeated pneumonia (n = 54); and group 4 had both swallowing abnormality and repeated pneumonia (n = 13). The follow-up period was as long as 2.2 years. Outcomes and causes of death were compared among the groups. RESULTS During the study, the overall mortality rate was higher in group 3 (24 deaths, 44.4%) and group 4 (9 deaths, 69.2%) than in group 1 (3 deaths, 4.3%, both p <.05). The annual mortality rate from pneumonia was also significantly higher in group 3 (21.2%) and group 4 (38.2%) than in group 1 (0.8%, p <.0001). The odds ratio for patients who subsequently died of pneumonia was 46.8 between groups 1 and 3. CONCLUSIONS The high sensitivity (.96) and specificity (.68) of swallowing ability indicate that the method is useful for identifying those persons at greatest risk for pneumonia and death after ischemic stroke.