Scintigraphy for evaluating early aspiration after oral feeding in patients receiving prolonged ventilation via tracheostomy

Objective: In tracheotomised patients the incidence of aspiration is difficult to determine because investigators often apply different criteria. In this study a scintigraphic method was used to visualise feeding aspiration directly and the results were compared with clinical evidence of aspiration. Design: Prospective study in difficult-to-wean patients with tracheostomy. Setting: Respiratory ICU. Patients and methods: The study population consisted of 62 consecutive patients (16 females, age: 64.1 +/- 11.1 years). All patients were tracheotomised and had previously been long-term ventilated in other ICUs due to weaning failure. The scintigraphic test was performed during spontaneous breathing. The standard nutrition consisted of a liquid, semi-liquid and solid meal which was labelled with 100 MBq 99mTc-human serum albumin. Measurements and results: Scintigraphic aspiration (SA) was defined as positive if radioactivity was detected in the bronchial system using a scintillation camera. Furthermore, aspiration was proven clinically (CA). CA and SA yielded identical results in 54 of the 62 patients [10 positive (16%) and in 44 negative (71 %)]. CA, but not SA, was seen in 4/62 (6.5 %) and SA, but: not CA, was found in 4/62 (6.5 %) patients. Conclusions: Our data re-emphasise that aspiration in tracheotomised patients is common (in our study approximately 30 %). The scintigraphic method failed to identify all tracheotomised patients with clinically significant aspiration; however, it did suggest that some patients had subclinical aspiration.