Limulus amoebocyte lysate and direct sampling methods for surveillance of operating nebulizers

Abstract

The Limulus amoebocyte lysate test for detection of endotoxin (Pyrogent; Mallinckrodt Chemical Co.) and the Easicult method (Orion Diagnostica) for detection of bacteria were compared with direct dilution sampling, a standardized technique for respiratory therapy surveillance previously developed in our laboratory. Tests of 206 reservoirs of nebulizers were done in three hospitals in Georgia. Forty-five percent of all reservoirs sampled were contaminated. Gram-negative, nonfermentative bacilli were the predominant contaminants. The results of the Limulus test and the Easicult system were in agreement with those of the direct dilution sampling tests approximately 84 and 90% of the time, respectively. Direct dilution of water samples onto blood agar plates was the most sensitive, reliable, and informative method for detecting viable bacteria. The Easicult and Limulus systems were sensitive enough to detect greater than or equal to 10(3) colony-forming units per ml. Positive Limulus tests and negative culture tests, reflecting detection of endotoxin but not of viable gram-negative bacteria, occurred in 20 of 206 (9.7%) instances. Positive cultures and negative Limulus tests were noted in 13 of 206 (6.8%) samplings. The Limulus test is a valuable procedure, for it can detect moderate-to-heavy microbial contamination within 1 h of testing and affords the opportunity to remove contaminated equipment from patients within minutes of a positive test result. These results demonstrate the potential value of the Easicult and Limulus tests for selective surveillance of operating nebulizers.