**4. Comparing OEP with spirometric operational volumes**

*These data suggest that pulmonary rehabilitation reduces dyspnea regardless of rib cage distortion*

Many COPD patients complain of severe dyspnea while performing simple daily-life activi‐ ties using their arms. The increased demand during simple arm elevation may play a role in the development of dyspnea and in the limitation that is frequently reported by these patients when performing activities involving their arms [29,30]. Unsupported arm exercise training (UAET) is increasingly recognized as an important component of pulmonary rehabilitation in these patients [31]. Although some studies have demonstrated improvement in unsupported arm exercise after UAET [32-34], suggesting that the test can be sensitive to changes in arm ex‐ ercise capacity, the impact of upper extremity training on arm exercise related-dyspnea and fa‐ tigue remains unclear [35-38] or undemonstrated [32,38-40]. Surprisingly, few studies [32,35,37-39] have investigated the effect of upper extremity training on ratings of perceived dyspnea by applying psychophysical methods, that is, the quantitative study of the relation‐ ship between stimuli and evoked conscious sensory responses. On this basis we have recently demonstrated that neither chest wall dynamic hyperinflation nor dyssynchronous breathing *per se* are the major contributors to dyspnea during unsupported arm exercise in COPD pa‐ tients [25]. Using the same approach we have recently tried to document the impact of arm training on arm exercise-related perceptions. The finding that before rehabilitation patients stop arm exercise namely because of arm symptoms, makes a case for the excessive effort felt by subjects being elicited by arm/torso afferent information (from the muscles performing the

These findings may explain why even a very small decrease in ventilatory demand, reflec‐ tive of a decrease in central motor output to ribcage/torso muscles, has a salutary effect on

OEP has also helped to clarify mechanisms by which some techniques of pulmonary rehabil‐ itation such as breathing retraining, namely "pursed lip breathing" (PLB), act in reducing the sensation of dyspnea. Bianchi *et al*. [42] hypothesized that the effect of PLB on breathless‐ ness relies on its deflationary effects on the chest wall. They found that patients exhibited a significant reduction in end-expiratory volume of the chest wall (Vcw,ee) and a significant increase in end-inspiratory volume of the chest wall in comparison with spontaneous breathing. In a stepwise multiple regression analysis, a decrease in end expiratory volume of

*These data indicate that by lengthening the expiratory time, PLB deflates the chest wall and reduces*

In a further paper Bianchi *et al.* [43] identified the reasons why some patients benefit from PLB while others do not. The OEP analysis of chest wall kinematics shows why not all patients with COPD obtain symptom relief from PLB at rest. The most severely affected patients who de‐ flate the chest wall during volitional PLB reported improvement in their sensation of breath‐

*and dynamic chest wall hyperinflation.*

476 Optoelectronics - Advanced Materials and Devices

excessive effort) conveyed to the motor-sensory cortex [25].

arm symptoms during arm training in patients with COPD [41].

the chest wall accounted for 27% of the variability in the Borg score.

lessness. This was not the case in the group who hyperinflated during PLB.

*dyspnea.*

OEP may provide complementary information on operational volumes to that provided by spirometry. Vogiatzis *et al.* [28] found a good relationship between changes in inspiratory capacity (ΔICpn) and changes in end expiratory chest wall volume (ΔVcw,ee). By contrast we have not found any significant relationship between the two measurements (Fig 3).

**Figure 3.** Plots of change in inspiratory capacity (IC) vs change in end-expiratory-chest wall-volumes (CWee) from rest to end exercise, before (closed circles) and after (triangles) pulmonary rehabilitation. Continuous line is the identity line.

The decrease in ICpn is much greater than the increase in Vcw,ee in most patients. The rea‐ sons for this discrepancy are probably due to: i) error measurements with the pneumotacho‐ graph possibly linked to leakage and elevation of temperature in the system, and to spirometric drift resulting in spurious increments or decrements in volume measurements; ii) spirometry measures the volume of the gas entering or leaving the lungs at the mouth, while OEP measures the volume of the trunk which includes changes in gas volume, gas compression and blood volume shifts [16]. Arguably, activity of the abdominal muscles pro‐ ducing various amounts of gas compression and blood shifts might account for the preva‐ lence of one method over the other. For instance, high gas compression and blood shift would result in a greater decrease in Vcw,ee than an increase in the next ICpn manoeuvre [44]. It has been postulated that OEP would not detect 89% of the reduction in inspiratory capacity measured with spirometry in some conditions [44].
