Body Plethysmography is Not Helpful for COPD Diagnosis, Determination of Severity, Phenotyping, nor Response to Therapy

Almost every hospital-based pulmonary function laboratory (PFT lab) uses a body plethysmograph (body box) to measure static lung volumes (total lung capacity [TLC], functional residual capacity [FRC], and residual volume [RV]) and airway resistance (Raw or specific Raw) for most adults referred to the PFT lab, including those at risk for COPD or patients with a diagnosis of COPD (1). However, despite this widespread practice, there is no evidence that after the results of a chest X-ray, pre- and post-bronchodilator (post-BD) spirometry, and diffusing capacity (DLCO) are considered, that the additional knowledge of lung volumes or airway resistance add clinically useful information for 1) the differential diagnosis of dyspnea in smokers at risk for COPD, 2) determination of COPD severity and prognosis, 3) determination of the COPD phenotype, or 4) response to COPD therapy (2).

1. 1) Current or former smokers, older than age 40, with chronic dyspnea are often referred to a pulmonary specialist to determine if their dyspnea is due to COPD, asthma, heart failure, or simply poor cardiovascular conditioning (often associated with obesity). Airflow limitation on spirometry makes asthma or COPD more likely than heart failure or obesity (which often show mild spirometric restriction, a low forced vital capacity (FVC) with a normal FEV₁ [the volume exhaled during the first second of a forced exhalation]/FVC). Substantial airflow limitation post-BD makes COPD more likely. In smokers with airway obstruction, the degree of BD responsiveness is usually not helpful for distinguishing asthma from COPD (3). Both asthma and COPD patients often have a BD response that fits the traditional American Thoracic Society (ATS)/European Respiratory Society (ERS) definition of significant BD responsiveness: FEV₁ or FVC increase of more than 12% or 0.2 L. Only when the patient has a large BD response (FEV₁ or FVC increase of more than 25% or 0.4 L), is asthma more likely than COPD. A low DLCO (after correction for any anemia) makes COPD more likely than asthma or heart failure (4).

Body box measurements do not add to the preceding tests for the differential diagnosis of dyspnea (5). Airway resistance is increased in both asthma and COPD, and Raw is normal with heart failure and obesity. Reference values for airway resistance are poorly established in comparison to spirometry, making body plethysmography less reliable than spirometry for detecting mild COPD. Due to air trapping and hyper-inflation, RV and FRC are elevated in patients with moderate-to-severe airway obstruction (and the vital capacity and inspiratory capacity are proportionally reduced). However, the degree of hyperinflation is directly associated with the degree of airway obstruction (as measured by percent predicted FEV₁), in both asthma and COPD (6). Thus, body box measurements do not help to differentiate asthma from COPD.

Cigarette smoking is the major cause of both COPD and heart failure. About 20% of patients with a diagnosis of COPD also have heart failure, which is often missed on the physical exam or chest X-ray (7). Since dyspnea due to heart failure responds very well to therapy (greatly reducing morbidity and mortality) it behooves pulmonologists to look for heart failure in patients with dyspnea, even for patients with post-BD airway obstruction (8,9). It is efficient to start with a simple B-naturetic protein (BNP) blood test (10). A high BNP value greatly increases the likeliness of heart failure, while a normal BNP value makes heart failure much less likely the etiology of the patient's dyspnea. An intermediate BNP value doesn't help much, so an echocardiogram or cardiac MRI should be obtained for such patients (7).

1. 2) COPD guidelines from professional societies, as well as the GOLD documents recommend using the percent predicted FEV₁ from spirometry, the degree of dyspnea on exertion, and other factors to determine COPD severity (and thus treatment). None suggest using the degree of hyperinflation, air trapping, or airway resistance. When heart failure or severe obesity are co-morbid conditions contributing to the patient's dyspnea, the FEV₁ and FVC are often reduced, exaggerating COPD severity based on percent predicted FEV₁ (9,11). Among pulmonary function tests, both spirometry and DLCO have been shown to be independent factors (statistically significant, but not clinically important) for predicting progression of COPD in adult smokers (12–15); however, no studies have considered body box measurements as prognostic factors.

2. 3) Identifying the patient's COPD phenotype helps to optimize therapy and improve patient outcomes (16). A low DLCO, blebs or bullae on chest X-rays, or low attenuation on lung CT scans make the emphysema phenotype more likely.

Up to one-third of adult smokers with airway obstruction present with a large BD response, suggest an asthmatic phenotype. These patients likely have “overlap syndrome” of both asthma and COPD and are more likely to respond better to inhaled corticosteroid therapy for asthma than to anticholinergic inhalers for COPD. Overlap syndrome typically only requires treatment for one of these diseases, not both of them (17). So, it is clinically important to detect the “hidden asthmatics” or COPD phenotype A (18) before prescribing a COPD inhaler for smokers with airway obstruction. Body plethysmography does not help to determine the asthma phenotype because hyper-inflation and air trapping occur with both asthma and COPD (6). PFTs are not helpful for identifying the COPD phenotypes of chronic bronchitis or frequent exacerbation.

Occasionally, patients with COPD also have a restrictive impairment, sometimes called a “mixed pattern.” Only about 10% of patients with a “mixed pattern” on spirometry (obstruction and a low FVC) have a low TLC when measured by body plethysmography (6, 19). These patients may have a restrictive disorder superimposed on asthma or COPD, but this could usually have been determined by examination of their chest X-ray. In the other 9 of 10 cases, the low FVC was due to hyperinflation (a high residual volume), caused either by asthma or COPD. The chest X-ray (or lung CT scan) will also show hyperinflation, but not infiltrates in these cases.

1. 4) The traditional measure of successful bronchodilator therapy is improvement in FEV₁ or lack of a rapid decline in FEV₁ over several years. However, successful bronchodilator therapy also reduces hyperinflation, as measured by a decrease in FRC and RV with an increase of inspiratory capacity and thus vital capacity (slow or forced). All of these changes in lung volumes occur in parallel, so studies have not yet been done that demonstrate that the presence, severity, or change in RV (as measured by body plethysmography) adds any additional clinically useful information to measurements of FVC, slow VC (SVC), or inspiratory capacity (IC) measured by spirometry. The FVC or IC are much easier to measure as an index of physiologic improvement when the airflow limitation and hyperinflation of severe COPD or asthma are successfully treated. The ATS/ERS 2005 PFT guidelines did not help with definitions of hyperinflation (perhaps a high RV/TLC), over-distention (perhaps a high TLC), trapped air volume (perhaps TLCbox minus VA), or “pendelluft” (20).

A DLCO test is more helpful than a body box test for patients with COPD. In adult patients with post-bronchodilator airway obstruction and a history of cigarette smoking, a low DLCO makes the emphysema phenotype of COPD highly likely (4). The emphysema phenotype can also be detected using a high-resolution lung CT scan (HRCT) (18), but a DLCO test is much less expensive and avoids the risks of radiation. Once a diagnosis is made, the percent predicted DLCO provides an objective index of disease severity and may guide the choice of initial therapy (16, 17). A DLCO below 40% predicted, or a decline in DLCO, is associated with increased morbidity and mortality, which should prompt more aggressive treatment, such as lung volume reduction (21). A low DLCO (as an index of emphysema) is an independent predictor of more rapid subsequent decline in lung function (13, 15) and excessive loss of exercise capacity (14).

In patients with severe COPD, the alveolar volume (VA) from the DLCO test is often smaller than the TLC measured by multi-breath techniques (19), but the TLC measured by a body box is often an over-estimate of the TLC measured by lung CT scan (22). However, the VA correlates with the non-emphysematous lung volume measured by a CT scan (23) and has been called “the effective lung volume” (24). In addition, a low VA is an independent predictor of mortality in COPD patients with co-morbid heart failure (25). Thus the VA may be as clinically useful in COPD as lung volumes measured by a body box.

Pulmonary subspecialists who see the majority of patients with severe COPD should therefore consider DLCO testing of such patients in their outpatient office, now that small instruments are available.

Declaration of Interest Statement

During the past 3 years, travel expenses for Paul Enright to give talks about pulmonary function testing have been reimbursed by professional societies, after they have received unrestricted educational grants from ndd Medical, a Swiss company that makes a portable instrument for measuring spirometry, DLCO, and multi-breath lung volumes. The author stubbornly refuses funding for consulting, even from big pharma. The author is responsible for the content and writing of this paper.