Abstract
158
Introduction: Salivagram was first described many decades ago by the physicians at The Children's Hospital of Philadelphia to detect possible lung aspiration in infants or young children. Its original protocol requires dynamic imaging for 1 hour which has high sensitivity. However, in select cases, it is difficult or impractical to have 60 minute image acquisition. SNMMI guideline recommended 30-60 minute dynamic imaging and some institutions therefore only acquire 30 minute images. However, there is insufficient data to show whether shortened imaging time is adequate to detect aspiration. The Purpose of this retrospective investigation is to determine whether a shortened period of image acquisition can be adopted without significantly reducing the sensitivity of the test in detecting aspiration.
Methods: The current salivagram protocol in our hospital requires 60 minute dynamic imaging. The images of a total of 398 patients (age 1 month to 9 years old) with positive salivagram were analyzed. The entire 60 minute dynamic images were divided to 6 periods, 10 minutes/period. The time of the onset of abnormal activity in the bronchi which was the evidence of aspiration in each patient was recorded and assigned into corresponding time period.
Results: Among all 398 patients with evidence of aspiration, tracheobronchial tree activity could be seen in the first 10 minutes of the dynamic imaging in 184 of them (46.2%, 184/398). The onset of the bronchi activity was seen between 10 and 20 minutes in 177 patients (44.5%, 177/398). A total 35 patients (8.8%, 35/398) had onset of abnormal tracheobronchial tree activity in the 3rd time period between 20-30 minutes. During the 4th time period between 30-40 minutes, the onset of the aspiration occurred only 2 patients (0.5%, 2/398). All patients had the onset of aspiration in the first 40 minutes of the dynamic imaging. Discussion: In our patient population, overwhelming majority of the aspiration could be visualized during the first 20 minutes of the dynamic images. Based on the data analyzed, 99.5% of aspiration onset occurred at the first 30 minutes and 100% at the first 40 minutes.
Conclusions: The originally-described 60 minute dynamic imaging protocol of salivagram can be safely shortened to 40 or even 30 minutes without significantly decreased chance of detecting aspiration. Prolonged imaging is unnecessary.