Accuracy of float testing for metered-dose inhaler canisters

Tina Penick Brock, Andrea M. Wessell, Dennis M. Williams, James F. Donohue

Research output: Contribution to journalArticleResearchpeer-review

12 Citations (Scopus)


OBJECTIVE: To characterize and evaluate canister floating patterns of three commercially available metered-dose inhalers (MDIs) with varying amounts of medication remaining. 

DESIGN: Four canisters each of three asthma medications were studied. MDIs were actuated every 30 seconds to 60 seconds, and canisters were weighed and floated at 100%, 75%, 66%, 50%, 33%, 25%, 10%, and 0% of remaining labeled actuations. Position of the canisters and percentage submersion in water were recorded. 

SETTING: Controlled laboratory. 

RESULTS: We observed differences among the products with regard to canister floating behavior at varying levels of fullness. All canisters were completely submerged with the nozzle up at two-thirds full and greater. The canisters remained nozzle-up and were submerged to varying levels at the half-full point. When observed at less than half full, canisters inverted and floated nozzle down. Positions of the canisters varied among products at less than half full. No canister was fully tilted when all labeled actuations were used. 

CONCLUSION: Float characteristics are product-specific and a function of canister size, design, content, and method of testing. Clinicians and asthma educators should not advise patients to use a float test to assess the amount of medication remaining in an MDI. Recommendations from the National Asthma Education and Prevention Program of the National Heart, Lung, and Blood Institute suggest that the only reliable method for determining the number of doses remaining in a canister is to subtract the number of doses used from the number available.

Original languageEnglish
Pages (from-to)582-586
Number of pages5
JournalJournal of the American Pharmaceutical Association
Issue number4
Publication statusPublished - Jul 2002
Externally publishedYes

Cite this