Sleep patterns in the intensive care unit and on the ward after acute myocardial infarction.

Twelve patients aged 33--70 years (mean 49.5) underwent nightly recordings in the ICU and subsequently on the ward following acute myocardial infarction. Sleep patterns were analyzed according to night after infarct and ICU versus ward environment. Significant differences in nocturnal sleep patterns from matched controls initially after infarction included greater wakefulness, low REM sleep per cent, long REM latency, fewer REM periods, more awakenings, more stage shifts and decreased sleep efficiency. The usual circadian variation in HR was absent, and there was an estimated 8--10 h of unrecorded daytime sleep, which together suggested a quite generalized disruption of biological rhythms. With time, there was loss of daytime sleep, lowered nocturnal wakefulness and increased REM sleep. Slow-wave sleep (sometimes with very long duration delta waves) increased above normal over post-infarction nights 3--9, and sleep was otherwise renormalized by post-infarction night 9. No sudden sleep changes occurred with transfer from ICU to ward. The altered sleep patterns appeared mainly attributable to infarction itself. Twelve nocturnal anginal attacks occurred. Ten began in NREM sleep and two in REM periods without particularly intense phasic activity. Post-infarction nocturnal angina therefore appears to differ in pathogenesis from angina outside this period, which usually occurs in REM sleep. ECG changes could occur during sleep before awakening with pain, and overall decrease in ECG amplitude sometimes accompanied angina. Most attacks (10 of 12) occurred on post-infarction nights 4 and 5, indicating that undetermined that undetermined factors produce a secondary period of heightened risk at that time.