Chronic exposure to diesel exhaust may cause small airway wall thickening without lumen narrowing: a quantitative computerized tomography study in Chinese diesel engine testers

Table 3 The dose-response between CCAM and wall area percent in all study subjects (n = 106)^a

Wall area percent	CCAM		N (DET)	Mean (95%CI)	Ratio (95%CI)	P
Wall area percent	Category	Range (%)	N (DET)	Mean (95%CI)	Ratio (95%CI)	P
6th generation (%)	1	0.24–1.33	21 (2)	49.7 (45.9, 53.7)	Ref
	2	1.33–1.92	21 (4)	54.3 (50.1, 58.9)	1.09 (0.98, 1.22)	0.102
	3	1.92–3.10	21 (9)	53.0 (49.0, 57.1)	1.07 (0.96, 1.19)	0.249
	4	3.10–4.70	21 (18)	52.0 (48.1, 56.3)	1.05 (0.94, 1.17)	0.391
	5	4.70–30.24	22 (20)	56.8 (52.5, 61.3)	1.14 (1.03, 1.27)	0.015
Trend test^b					1.02 (1.00, 1.05)	0.063
9th generation (%)	1	0.24–1.33	21 (2)	39.4 (35.1, 44.3)	Ref
	2	1.33–1.92	21 (4)	45.3 (40.2, 51.1)	1.15 (0.98, 1.35)	0.085
	3	1.92–3.10	21 (9)	43.9 (39.2, 49.1)	1.11 (0.95, 1.31)	0.195
	4	3.10–4.70	21 (18)	43.0 (38.3, 48.3)	1.09 (0.93, 1.28)	0.287
	5	4.70–30.24	22 (20)	45.1 (40.2, 50.6)	1.14 (0.98, 1.34)	0.098
Trend test^b					1.02 (0.99, 1.06)	0.239

Definition of abbreviations: CCAM carbon content in airway macrophage, CI confidence interval, DET diesel engine tester, LsMean least square mean, Ref reference
^a Linear mixed effects model assessed differences of natural log transformed 6th and 9th wall area percent between categories 2–5 and the reference group (category 1) with adjustment of age, BMI, smoking history, CT reconstruction method, and lung lobes. Mean and 95%CI are exponentials of least square means and 95%CIs calculated based on natural log transformed data. Ratio = e^β, 95%CI = e^{(β ± 1.96 × Se)}
^b Linear mixed effects model assessed dose – response relation between CCAM category coded as 0, 1, 2, 3, and 4 and natural log transformed 6th and 9th wall area percent with adjustment for covariates listed above

ISSN: 1743-8977