Table 2:
Multivariable zero-inflated negative binomial regression analysis of the association between quantitative CT biomarkers and severe exacerbations during 3-year follow-up in the SPIROMICS cohort
Count component | Inflation component | |||
---|---|---|---|---|
Incidence rate ratio (95% CI) | p value | Odds ratio (95% CI) | p value | |
Age | 0·991 (0·974–1·008) | 0·281 | 0·942 (0·885–1·002) | 0·058 |
Sex | 1·259 (0·972–1·632) | 0·081 | 1·277 (0·586–2·781) | 0·539 |
Race (White vs other) | 0·901 (0·684–1·187) | 0·459 | 2·240 (0·768–6·530) | 0·14 |
Baseline FEV1 (% predicted) | 0·986 (0·978–0·993) | 0·0002 | 1·007 (0·982–1·033) | 0·579 |
Smoking status | 0·907 (0·658–1·251) | 0·553 | 0·577 (0·210–1·591) | 0·288 |
BMI, kg/m2 | 1·011 (0·986–1·037) | 0·376 | 0·982 (0·911–1·058) | 0·628 |
History of exacerbations* | 1·137 (1·028–1·259) | 0·013 | 0·132 (0·027–0·640) | 0·012 |
SGRQ† | 1·006 (0·997–1·015) | 0·204 | 0·921 (0·888–0·955) | <0·0001 |
CT density gradient | 0·979 (0·939–1·020) | 0·311 | 0·758 (0·595–0·967) | 0·026 |
Wall area, % | 0·019 (0·0002–1·560) | 0·078 | 0·0003 (0–1654·17) | 0·312 |
Pi10 | 3·174 (0·942–10·700) | 0·062 | 1·827 (0·040–84·027) | 0·758 |
1535 participants were included in the analysis (1204 with no exacerbations and 331 with at least one acute episode in 3 years) across GOLD stages (0–4), including individuals who never smoked. The zero-inflated negative binomial regression model was adjusted for age, sex, postbronchodilator FEV1, smoking status, BMI, history of exacerbations, and symptom burden score quantified by SGRQ. SPIROMICS=Subpopulations and Intermediate Outcome Measures In COPD Study. GOLD=Global Initiative for Chronic Obstructive Lung Disease. FEV1=forced expiratory volume in 1 s. SGRQ=St George’s Respiratory Questionnaire. Pi10=square root of airway wall area of a hypothetical airway with an internal perimeter of 10 mm.
Exacerbation history included count of all exacerbation episodes within the year preceding enrolment.
The score ranges between 0 and 100, higher scores indicating increased symptom burden.