The stable microbubble (SM) test on gastric aspirate obtained at birth proved useful in identifying infants who would develop respiratory distress syndrome (RDS). This test involves only the count of stable microbubbles of < or = 15 microns in diameter. Larger bubbles (> 15 microns in diameter) are not necessary for the test and may interfere with stable microbubble counting. The aims of the present study were to determine: (i) if larger bubbles could be selectively removed by adding ethanol, a potent bubble breaker; and (ii) if the predictive value of this modified test, the ethanol resistive microbubble (ERM) test, on the development of RDS was similar to that of the SM test. Varying amounts of different concentrations of ethanol-water solutions were added to the top of the bubble crop generated by the SM test procedure, and the mean counts of stable microbubbles and larger bubbles in five regions were calculated. A volume of 10 microL of 47.5% ethanol was effective in defoaming larger bubbles generated by the SM test procedure without altering the stable microbubble counts. When concurrently performed on 43 samples of gastric aspirate obtained at birth from infants of less than 35 weeks gestation, the RDS predictive value of the ERM test was similar to that of the SM test. It was concluded that the ERM test may serve as an alternative to the SM test.