Abstract
To analyse the effects of apnoea and bradycardia on the oxygen saturation (SaO2) of preterm infants and to make recommendations for apnoea alarm limits, polygraphic recordings were made on 89 occasions of 27 preterm infants; 1029 apnoeic episodes were analysed. Reduction in SaO2 was positively correlated with duration of apnoea, but the scatter of results was such that reductions in SaO2 of up to 40% occurred with apnoeas of less than 10 seconds duration. The median initial SaO2 was significantly lower in those episodes that resulted in bradycardia (92% compared with 95%), and there was also a significantly greater reduction in median SaO2 (9% compared with 5%). This study illustrates the difficulty of setting alarm limits for the detection of apnoea. We suggest that rather than simply detecting apnoea it is more appropriate to monitor heart rate and SaO2 in infants with recurrent apnoea.