Introduction and objectives Abnormalities in indices of gas mixing derived from multiple breath nitrogen washouts have been demonstrated in smokers without spirometric evidence of airflow obstruction, which improved on smoking cessation. We report initial data from a study investigating the ability of a simpler measure of gas mixing, Lung Clearance Index (LCI), derived from multiple breath washouts using sulphur hexafluoride as tracer gas and a modified Innocor gas analyser, to detect early airway changes in smokers with normal spirometry.

Methods Current cigarette smokers with a smoking history of over 10 pack years and no known cardiac or respiratory disease were recruited from smoking cessation clinics. Spirometry was performed to ERS standards before and after salbutamol. Participants with post-bronchodilator FEV₁<80% predicted, FEV₁/FVC<0.7, or a significant bronchodilator response were excluded from analysis. St George's Respiratory Questionnaire (SGRQ) was completed. LCI reported is the mean of at least two technically acceptable repeat measurements.

Results 17 participants remained after exclusions, mean age 44 years (range 31–57) and mean smoking history of 25 pack years (range 11–60). Mean (SD) post-bronchodilator FEV₁ and FEF_25–75 was 101 (12) and 80 (23)% predicted respectively. Only one participant had FEF_25–75<60%. There was a moderate negative correlation between smoking history and FEF_25–75 (r=−0.51, p=0.037) but not between smoking history and FEV₁. Mean (SD) LCI was 7.7 (0.98) with mean (SD) intra-visit coefficient of variation of 3.7 (2.5)%. Eight participants had LCI>7.5 (95% CI for LCI in normal subjects 5.9–7.5), suggesting impaired lung gas mixing. There was a negative correlation between LCI and FEV₁ (r=−0.55, p=0.02) and between LCI and FEF_25–75 (r=−0.66, p=0.004) but no significant correlation between LCI and smoking history or total SGRQ score. Mean (SD) total SGRQ score was 10.9 (7.5), 12 participants scoring over 7 (95% CI for total SGRQ score in normal subjects 5–7). Total SGRQ did not correlate with smoking history, LCI, FEV₁ or FEF_25–75.

Conclusions These data support the hypothesis that LCI is a sensitive marker of early airway changes in smokers with normal FEV₁ and FEF_25–75. The effects of smoking cessation on this measure are currently being investigated.