Tube Check: A Plain Question with Quality Implications
What benchmarks are indications of successful nasogastric tube placement?
It’s 10 p.m. Your workflow begins to flood with requests from the emergency department. The ICU intern inserts a nasogastric tube into the patient you diagnosed 25 minutes prior with transcortical infarction. The film transmits to PACS, “Indication: Tube check.” Opening the film, you examine the thin, white line coursing midline down the spinal vertebra. As your eyes reach the diaphragm, the tube deviates slightly left; the side port overlies the stomach; and you open the report to enter a dictation. “NG tube with side-port and tip overlying the region of the stomach.” The report is signed; and another case drops from the list. It’s 10:01 p.m.
All too familiar, especially during a busy call shift, the inpatient teams order plain films of the thoracoabdominal cavity to confirm placement of tubes, lines, and drains. Some patients are subjected to repeated x-rays to confirm placement of difficult insertions. What metrics can be investigated to measure the effectiveness of both the radiologist and the referring physician in this scenario?
In the June 2017 issue of the JACR®, one team highlights this question, and sets out to measure the quality and safety of nasogastric tube (NGT) insertions at their institution. Investigators from Montreal Neurological Hospital and Institute in Quebec, Canada, embraced the often-unasked question: what underlies cases requiring multiple radiograph studies to confirm proper NGT position?
This retrospective study used the outcome measure of number of radiographic studies per single insertion procedure. The lower the ratio, the more effective and safe their protocol, as interpreted by the investigators. After applying exclusion criteria to an initial 407 radiographs, 100 cases met sufficient criteria to be analyzed. While relatively small, the stratification of men to women was equal with a mean age of 69 years.
Exclusion criteria screened out patients requiring multiple NGT insertions per admission. The rationale for screening out this subset was the bias introduced from patient cooperation, neurologic condition, and traumatic previous attempts. Essentially, the two top diagnoses of included patients consisted of stroke and neuromuscular disease.
Overall, the research team used the outcome measure of radiographs acquired per single insertion event to conclude that a ratio of 1:1 translates to safe, effective NGT insertions. There is no mention of how physicians are trained to insert NG tubes; nor is their insight into why their institution is so successful. They simply issue the outcome measure with data gathered from their institution.
One point to consider: The study does not account for the number of attempts to place an NGT prior to the first radiograph. Essentially, if deemed malpositioned by the proceduralist, the tube may have been removed and reinserted prior to the initial scan. This introduces a Type I error; the team rejects the hypothesis that NGT insertions are not safe at their institution. Thus, the published results may be falsely positive.
The genius of quality improvement rests in the reproducibility of results through the continuous Plan Do Study Act (PDSA) cycle. Through continuous PDSA, this outcome measure of radiographs per NGT insertion can be assessed and analyzed. The initiative at Montreal Neurological Hospital and Institute should inspire health systems seeking to reproduce outcomes.
By Michael A. Chorney, diagnostic radiology resident, Pennsylvania Hospital, University of Pennsylvania Health System, Philadelphia