Implications for practice and research

• There is insufficient evidence to support ultrasonography as a method to rule out an improperly positioned nasogastric tube.

• Large controlled studies are needed to determine circumstances under which ultrasonography is most likely to predict correct nasogastric tube placement.

Context

Nasogastric tube insertion is a commonly performed procedure in patients of all ages in a variety of settings. At present, a properly obtained and interpreted X-ray is the standard for determining correct tube placement. To reduce radiation risk and cost, investigators continue to search for bedside testing methods that approach the accuracy of radiography. The ideal bedside test would have high sensitivity and specificity and be easy to perform with minimal equipment. Specificity is especially important to assure that the method could detect an improperly positioned tube. A dreaded event is instillation of feedings, medications or diagnostic dyes via a tube inadvertently positioned in the respiratory tract. The study reported by Lin et al 1 evaluated the accuracy of ultrasonography in determining nasogastric tube placement.

Findings

Following the identification of 10 102 references through electronic searches, 85 full-text articles were retrieved for assessment; of these, five were included in the analysis. Three of the studies were conducted in France, one in Korea and one in China. A total of 420 cases from the five studies were available for analyses. Four of the studies were published in refereed journals between 2012 and 2014; the fifth was a dissertation reported in 2013. A forest plot was used to report the number of true positives (n=357), true negatives (n=29), false positives (n=5) and false negatives (n=29). The investigators were unable to use a funnel plot to explore reporting bias because of insufficient studies.

Commentary

The study by Lin et al 1 is an important addition to the literature in that it provides a review of the accuracy of ultrasonography when used in a variety of patient care settings to test tube placement. However, missing from the manuscript is a discussion of the false-positive cases identified in the study. That is, ultrasonography indicated that 5 of the 420 tubes were in the stomach when actually they were elsewhere (sites undefined). While a false-positive incidence of 1.19% is small, it has great clinical significance. For example, if one or more of these tubes were positioned in the respiratory tract, introduction of feedings or medications could have caused serious respiratory problems. Air or saline was injected via the tubes in four of the five studies. If saline instillation is a requisite for testing tube placement by ultrasonography, this method should not be used for blindly inserted nasogastric tubes that could conceivably be in the airway.

Three of the studies included in the analysis used neck and oesophageal probes in addition to a gastric probe; this is a reasonable approach since it could enhance detection of a tube that ends in the oesophagus (an important consideration because introducing feedings via a tube ending in the oesophagus greatly increases risk for aspiration). As with any placement test, skill level of the observer is an important element of the method’s potential for success. Proficiency in using ultrasonography was not addressed in the five studies reviewed by Lin et al. 1

Ultrasonography equipment is not typically available on patient care units. Further, an undetermined level of training on use of the equipment would be required before bedside nurses could successfully use ultrasonography. These reasons, added to the questionable sensitivity and specificity of ultrasonography, dampen enthusiasm for its use in routine clinical settings to test placement of nasogastric tubes.