The radiopacity of ingested medications.

We investigated prospectively the radiopacity of 312 pills found on a university hospital formulary by first radiographing them through 15 and 25 cm of water to duplicate the radiodensity of the human body. The pills that were radiodense through water were studied in a human cadaver model, and their densities were quantified by computed tomography (CT). Thirty-five of 312 pills were radiopaque in 15 cm or more of water, and 23 of these pills were radiopaque on a plain radiograph when placed in the stomach of a cadaver. Common mnemonics used to identify radiopaque pills were found to be incomplete and inadequate. Chloral hydrate, iron-containing preparations, calcium carbonate, iodinated compounds, acetazolamide, busulfan, and potassium preparations were consistently radiopaque. Antihistamines, phenothiazines, and tricyclic antidepressants demonstrated varying radiopacity. There was varying radiopacity among the same medications made by different manufacturers. The presence of an enteric coating did not assure that the pill would be radiopaque. Merely radiographing a pill that has been placed on a standard radiograph cassette will make pills that are actually radiolucent in the body appear radiopaque. This test cannot be used to predict radiopacity in vivo. Visibility when radiographed through 15 cm or more of water and a CT radiodensity of more than 1,300 Hounsfield units are predictors of the radiopacity of a pill in the stomach of a cadaver model on a standard KUB radiograph. Variables, such as the size of the patient, the arrangement of pills in the stomach, air contrasting a pill, and the specific composition of the enteric coating or the pillmatrix, affect the radiodensity of pills.(ABSTRACT TRUNCATED AT 250 WORDS)