OBJECTIVE: To study the association between the frequency of videofluoroscopic modified barium swallow (VMBS) studies and the incidence of pneumonia in stroke rehabilitation patients. DESIGN: Retrospective comparative study. SUBJECTS AND SETTING: Five hundred sixty-three consecutive stroke patients admitted to one hospital rehabilitation unit in London, Ontario, Canada were compared with 461 consecutive stroke patients admitted to another hospital rehabilitation unit in the same city. INTERVENTIONS: The number of initial and total VMBS studies and the timing from stroke onset to initial VMBS studies. MAIN OUTCOME MEASURE: Incidence of pneumonia. RESULTS: At the first hospital, 146 patients (25.9%) had 232 total VMBS studies performed, whereas at the second hospital 57 patients (12.4%) had 73 total studies (p<.001). For the first 15 days there was no significant difference in the number of initial VMBS studies performed (8.2% vs. 9.2%). There was a marked difference in the number of initial VMBS studies performed after 15 days (17.2% vs. 2.0%, p<.0001). The difference between the hospitals in the number of VMBS studies in patients with brain stem strokes was not statistically significant, but for patients with hemispheric stroke, the difference was statistically significant. Pneumonia developed in 12 patients at the first hospital (2.1%) and 10 patients at the second hospital (2.2%), a difference that was not significant. CONCLUSIONS: The more frequent use of VMBS beyond 15 days after stroke was not associated with a change in the incidence of pneumonia among hemispheric stroke rehabilitation patients, assuming the two units were otherwise similar.
OBJECTIVE: To study the association between the frequency of videofluoroscopic modified barium swallow (VMBS) studies and the incidence of pneumonia in stroke rehabilitation patients. DESIGN: Retrospective comparative study. SUBJECTS AND SETTING: Five hundred sixty-three consecutive strokepatients admitted to one hospital rehabilitation unit in London, Ontario, Canada were compared with 461 consecutive strokepatients admitted to another hospital rehabilitation unit in the same city. INTERVENTIONS: The number of initial and total VMBS studies and the timing from stroke onset to initial VMBS studies. MAIN OUTCOME MEASURE: Incidence of pneumonia. RESULTS: At the first hospital, 146 patients (25.9%) had 232 total VMBS studies performed, whereas at the second hospital 57 patients (12.4%) had 73 total studies (p<.001). For the first 15 days there was no significant difference in the number of initial VMBS studies performed (8.2% vs. 9.2%). There was a marked difference in the number of initial VMBS studies performed after 15 days (17.2% vs. 2.0%, p<.0001). The difference between the hospitals in the number of VMBS studies in patients with brain stem strokes was not statistically significant, but for patients with hemispheric stroke, the difference was statistically significant. Pneumonia developed in 12 patients at the first hospital (2.1%) and 10 patients at the second hospital (2.2%), a difference that was not significant. CONCLUSIONS: The more frequent use of VMBS beyond 15 days after stroke was not associated with a change in the incidence of pneumonia among hemispheric stroke rehabilitation patients, assuming the two units were otherwise similar.