Is alveolar carbon monoxide an unreliable index of carboxyhaemoglobin changes during smoking in man?

1. We measured alveolar carbon monoxide (CO) after a 20 s breath-holding period and carboxyhaemoglobin both before and after smoking a cigarette on 500 occasions (101 individuals). The two measurements were closely correlated but there was a marked difference in the change or 'boost' after smoking one cigarette. The mean relative boosts ([post value--pre value]/[pre+post]/2) for alveolar CO and carboxyhaemoglobin were 7.7% and 20.3%, while negative boosts (fall rather than the expected rise) were seen in 103 of 500 and three of 500 occasions respectively. In 140 studies a third alveolar CO reading taken 5 min later was slightly larger, but the difference was insignificant. 2. In seven subjects where the carboxyhaemoglobin level was raised by breathing a 2% CO gas mixture, the alveolar CO and carboxyhaemoglobin boosts were similar (71.7% and 75.2% respectively), and they fell sharply subsequently rather than increasing further as occurred after smoking. 3. We conclude that alveolar CO measurements give a useful estimate of carboxyhaemoglobin level if the subject has not smoked for at least half an hour but that measurements of alveolar CO boost are useless since the act of smoking interferes with alveolar sampling. We postulate that cigarette smoking induces a transient change in pulmonary gas exchange.