Visit-to-visit FEV1-variation and Mortality in WTC Exposed FDNY Rescue/Recovery Workers.

Abstract

Rationale: FEV₁ and its longitudinal change are mortality risk-factors. Visit-to-visit-FEV₁-variation is a risk-factor for death in cystic fibrosis but has not been studied in other cohorts.

Objective: Assess if longitudinal visit-to-visit-FEV₁-variation is a mortality risk-factor in World Trade Center (WTC) exposed FDNY rescue/recovery workers.

Methods: Linear mixed-effects regression of all post-9/11/2001 FEV₁ measurements defined time-effect on longitudinal-FEV₁-decline (FEV₁-slope) and its standard error (visit-to-visit-FEV₁-variation). Cox proportional hazards and logistic models adjusted for age and smoking assessed the association between FEV₁ related risk-factors and mortality. Receiver operating characteristic area under the curve (ROC-AUC) assessed predictive model performance.

Measurements and main results: Among 11,745 workers with ≥3 FEV₁ measurements, 575 (4.9%) died. When all FEV₁-related risk-factors were combined, each 5 mL/year increase in visit-to-visit-FEV₁-variation increased mortality 2.1-fold (HR=2.14; 95%CI=1.84-2.48); each 10%predicted reduction in the last-longitudinal-FEV₁ increased mortality 15% (HR=1.15; 95%CI=1.09-1.21), but each 10ml/year longitudinal-FEV₁-decline was not associated with mortality (HR=1.04; 95%CI=0.99-1.10). The ROC-AUC of a fully adjusted multivariable cumulative mortality model was 0.82 (95%CI=0.80-0.84); for unadjusted visit-to-visit-FEV₁-variation, AUC was 0.80 (95%CI=0.78-0.82); for last-longitudinal-FEV₁ AUC was 0.61 (95%CI=0.59-0.64) and for longitudinal-FEV₁-decline AUC was 0.58 (95%CI=0.56-0.61). In the 1,988/11,745(16.9%) with high-WTC-exposure defined as arrival at the WTC-site before noon on 9/11/2001, the risk of high-visit-to-visit-FEV₁-variation (top-quartile, ≥10.35 ml/year) increased 25% (OR=1.25; 95%CI=1.12-1.40).

Conclusions: Visit-to-visit-FEV₁-variation is a mortality risk-factor in FDNY rescue and recovery workers with greater accuracy for predicting cumulative mortality than either last-longitudinal-FEV₁ or longitudinal-FEV₁-decline. Further investigation in other cohorts is needed to assess the generalizability of this rarely studied mortality risk-factor.