Is This Really a New START in Chronic Obstructive Pulmonary Disease?

To the Editor:

Airflow obstruction in postbronchodilator spirometry (i.e., a low FEV₁:FVC ratio) is integral to the definition of chronic obstructive pulmonary disease (COPD). The severity of obstruction, however, is usually gauged by the decrease in the numerator alone (e.g., percentage predicted FEV₁ in the Global Initiative for Chronic Obstructive Lung Disease scheme) (1). Bhatt and colleagues recently recognized this paradox, proposing a new classification on the basis of FEV₁:FVC (staging of airflow obstruction by ratio [STAR]) (2). Although their results in three large cohorts are encouraging, the clinical use of FEV₁:FVC to quantify the severity of airflow obstruction in individual patients with COPD requires some caution.

FEV₁ is a reasonably good metric of the “fast” component of lung emptying, that is, that related to the airflow across the middle and large(r) airways. In contrast, the “slow” component is strongly influenced by the functional characteristics of the small(er) airways, being critical to determining the residual volume (RV). Air trapping (high RV) worsens in tandem with disease severity because of either progressive small airway disease or loss of alveolar attachments caused by widespread emphysema (3).

In this context, it is rather axiomatic that unless TLC increases (hyperinflation) in tandem with, or out of proportion to, RV, FVC would tend to decrease as COPD evolves to its later stages, increasing (or at least stabilizing) the FEV₁:FVC ratio despite disease worsening. For instance, two patients with similar TLC and FEV₁ impairment may show widely different RVs and RV:TLC ratios (respectively, 140% and 0.50 in patient A and 200% and 0.65 in patient B). The consequences will be a substantially lower FVC and a higher FEV₁:FVC ratio (and thus a lower STAR stage) for patient B, who shows worse air trapping. The authors report that within each Global Initiative for Chronic Obstructive Lung Disease stage, increasing STAR stage was associated “with monotonic increases in hyperinflation and air trapping.” In fact, it seems more appropriate to state “with monotonic increases in TLC and RV,” as their sample with lung volume measurements (the University of Alabama cohort) showed an unusually low burden of air trapping (RV 78.9 ± 19.7%) and hyperinflation (TLC 76.3 ± 36.9%). In any case, as RV:TLC ratios and their impact on FVC and FEV₁:FVC were not shown, it is still possible that STAR confers an advantage to individual patients with higher RVs at a given TLC (or lower TLCs at a given RV), both decreasing FVC and increasing FEV₁:FVC at iso-FEV₁. The corollary is that despite proving useful to discriminate a group of subjects at greater risk of a negative outcome (4), STAR may end up underestimating the severity of airflow obstruction in individual patients showing worse air trapping and more advanced COPD (i.e., those more frequently seen by pulmonologists).

The authors correctly argue that coexistent restriction may relatively worsen the severity of obstruction as indicated by percentage predicted FEV₁, as part of the FEV₁ variance can be explained by FVC. Conversely, it might be contended that by increasing FEV₁:FVC, restriction would underestimate the severity of obstruction as indicated by FEV₁:FVC.

Overestimation of disease severity by FEV₁:FVC may occur in the subset of patients with only mildly reduced FEV₁ who present with higher than expected FVC because of an undue increase in TLC (3). The causes of their lung overdistension beyond that anticipated by the emphysema burden remain unknown, being probably more common in subjects born with particularly compliant lungs. Whether dysanapsis also contributes to this pattern remains to be demonstrated.

The FEV₁:FVC conundrum in COPD (5) is an excellent example of the enduring potential of pulmonary function tests beyond simple spirometry to deeply phenotype patients with such a heterogeneous disease. Exactly because of the multifaceted nature of COPD, every attempt to encapsulate the severity of dysfunction in a single domain (such as airflow obstruction) has faced limitations when applied to the care of individual subjects (6). Perhaps future research will bring us some sort of multidimensional index geared to jointly gradate the abnormalities in gas transfer (e.g., transfer factor), airflow (e.g., percentage predicted FEV₁ or FEV₁:FVC, whichever is lower), and lung volumes (e.g., percentage predicted inspiratory capacity or ratio of inspiratory capacity to TLC) to be used with metrics of emphysema extension and symptom severity (dyspnea burden). Hopefully, such “GALES” would bring fresh wind to our sails toward a more holistic understanding of this fascinating disease.