Report on the development of a procedure to prevent placement of feeding tubes into the lungs using end-tidal CO2 measurements

Objective: To determine the accuracy of a technique using capnography to pr event inadvertent placement of small-bore feeding tubes and Salem sump tube s into the lungs. Setting: Twelve-bed medical intensive care unit (MICU) in a 557-bed teachin g hospital. Patients: A total of 25 ventilated adult MICU patients were studied-5 in ph ase 1 and 20 in phase 2. Design: Phase 1 tested the ability of the end-tidal CO2 (ETCO2) monitor to detect flow (and thus accurately detect CO2) through small-bore feeding tub es. A small-bore feeding tube, with stylet in place, was placed 5 cm throug h the top of the tracheostomy tube ventilator adapter in five consecutive p atients. The distal end of the feeding tube was attached to the ETCO2 monit or. The ETCO2 level and waveform were assessed and recorded. Because CO2 wa veforms were successfully detected, a convenience sample of 20 adult MICU p atients who were having feeding tubes placed (13 Salem sump tubes, 7 small- bore feeding tubes) was then studied. The technique consisted of attaching the ETCO2 monitor to the tubes and observing the ETCO2 waveform throughout placement. Results: The study hypothesis was supported. Of the seven small-bore feedin g tubes tested, all were successfully placed on initial insertion. Placemen t was confirmed by absence of an ETCO2 waveform and by radiograph. Of the 1 3 Salem sump tubes, 9 were placed successfully on first attempt and confirm ed by absence of CO2 and by air bolus and aspiration of stomach contents. E TCO2 waveforms were detected with insertion of four of the Salem sump tubes ; the tubes were immediately withdrawn, and placement was reattempted until successful. Conclusions: The technique described is a simple, cost-effective method of assuring accurate gastric tube placement in critically ill patients.