In patients with chronic obstructive pulmonary disease (COPD), exercise training is a core component of pulmonary rehabilitation, supported by high-level evidence.1 The 6-min walk test (6MWT) has been adopted as standard procedure for exercise capacity assessment in patients with pulmonary diseases.2 It is a simple, widely used and reliable test, useful for evaluating the functional status and prognosis of patients with a wide variety of diseases and the effects of several interventions, such as rehabilitation, pharmacological therapy and oxygen supplementation.3–5

Patients with COPD tend to stop exercising because of dyspnea rather than leg fatigue.5,6 Their physical performance may be reduced by a significant ventilatory limitation to exercise, particularly if peripheral muscle strength is not impaired.7–9 Consequently, adjunctive strategies to improve exercise tolerance have been proposed, such as non-invasive ventilation10 and the application of positive expiratory pressure (PEP).11

PEP is applied at the mouth in order to counterbalance the dynamic compression of the airways that leads to exercise-induced pulmonary hyperinflation. In some patients with COPD, reduced elastic recoil and the presence of expiratory flow limitation lead to pulmonary hyperinflation. Patients may present increased residual volume at rest (static hyperinflation) and/or may develop hyperinflation under strain (dynamic hyperinflation), as the increase of the respiratory rate shortens the expiratory time.12 Furthermore, airway dynamic compression may occur during expiration, especially at high flow rates. The equal pressure point (EPP) is shifted peripherally, resulting in an earlier closure of the peripheral airways with consequent air trapping. This phenomenon causes a gradual shift of the tidal volume (Vt) to higher levels of functional residual capacity (FRC) and a progressive decrease in inspiratory capacity (IC). Dynamic pulmonary hyperinflation can be quantified by measuring that decrease in IC.13

The application of PEP increases intraluminal airways pressure and moves the EPP back to the upper airways, reducing dynamic compression and subsequently limiting pulmonary hyperinflation.11,14 A similar effect is obtained by pursed lip breathing (PLB), which can be considered spontaneous PEP.15

In COPD patients, some studies investigated the effects of PEP on strength, endurance, lung volumes or walking exercise. Padkao et al.14 showed that the use of a flow-dependent conical-PEP lengthens isotonic quadriceps strength exercise periods. It also produces an increase in slow vital capacity (SVC) and in IC (+200ml, P=.05), with decreased pulmonary hyperinflation. Monteiro et al.16 applied PEP using an oronasal mask after a submaximal treadmill exercise in moderate-to-severe COPD patients, reporting an increase in IC after exercise. Furthermore, Martin and Davenport17 demonstrated that extrinsic 10cmH2O PEP reduced post-exercise dyspnea in COPD patients. More recently, in severe COPD patients, Nicolini et al.11 found that an adjunctive 5cmH2O PEP, administered via mouthpiece during the execution of a 6MWT, significantly improved the distance covered, oxygen saturation, and heart rate.

However, contrasting results were reported by Wibmer et al.,18 who applied flow-dependent PEP ranging from 10 to 20cmH2O during the 6MWT in patients with stable mild-to-severe COPD. They showed that PEP reduced FRC and the residual volume (RV) after exercise, but the PEP group walked 30m less than controls. Despite these reports, no studies have described the optimal level of PEP that could improve exercise tolerance.

In this context, the main aim of our study was to investigate the effect of two different levels of PEP valve (1cmH2O vs 10cmH2O) on the distance covered during a 6MWT (6MWD), compared to the basal 6MWT performed without PEP, in patients with severe COPD (FEV1<50%). Secondary outcomes were the evaluation of PEP effects on pulmonary function, symptoms and vital parameters immediately after the 6MWT.

This prospective, single-blind, randomized crossover study was performed at the Respiratory Rehabilitation Unit of Sestri Levante Hospital, Italy and the Pulmonary Rehabilitation Unit of Fondazione S. Maugeri IRCCS, Lumezzane, Brescia Italy, from January 2013 to October 2013.

Seventy-two patients were initially recruited. Fifteen subjects were excluded because they were unable to perform the 6MWT and 7 subjects refused to participate, so the final sample comprised 50 subjects. Table 1 describes the clinical and anthropometrical characteristics of the study patients. The patients were relatively elderly, predominantly male, and had severe COPD. A quarter of them were receiving long-term oxygen therapy due to associated respiratory failure. All patients had also severe static pulmonary hyperinflation. All 50 patients completed the study. Patient disposition is described in Fig. 1.

Fig. 1.

Patient disposition.

(0.1MB).

F-test for primary outcome (6MWD) shows a significant difference in results among different PEP levels and baseline (F-test 0.02). Post-hoc analysis shows that the use of PEP1 and PEP10 produced an increase in walking distance (PEP1 P<.002 and PEP10 P<.018), compared to baseline. There was no significant difference between PEP1 and PEP10. Fig. 2 shows the box plots of 6MWD.

Fig. 2.

Box plot of 6MWD without PEP (baseline), and using PEP1 and PEP10 cmH2O.

(0.08MB).

All changes (end of 6MWT−baseline) in secondary outcomes are reported in Table 2. We observed a tendency to decreased HR and increased RR in both PEP1 and PEP10 condition, compared to baseline, which did not reach statistical significance. No significant difference was found between PEP1 and PEP10 in changes in pulmonary function, symptoms and physiological variables.

A reduction in IC of 10% and/or >150ml, compared to baseline was defined as dynamic hyperinflation.22,23 Ten subjects showed dynamic hyperinflation and 40 did not. In the 40 subjects without dynamic hyperinflation, we observed a significant change in the distance covered (Δ meters) if 1cmH2O PEP was applied. No significant change was found if 10cmH2O was applied. These results appear in Table 3.

In this randomized crossover trial, conducted in patients with severe COPD, we found that the application of 1cmH2O PEP during the execution of 6MWT can improve the distance walked to the same extent as the application of 10cmH2O PEP, without affecting the cardiovascular response to exercise. We did not find any effect of PEP on dyspnea at the end of exercise and post-exertion lung dynamic volumes, particularly IC, SVC, FVC and ERV. Moreover, we found that the improvement of 6MWD is greater in patients with no or low dynamic hyperinflation when a low PEP is applied.

So far, only two other crossover randomized trials have investigated the effects of a PEP device on 6MWD, with contrasting results.

In 2013, Nicolini et al.11 found that a threshold PEP of 5cmH2O improved the distance walked during the 6MWT in patients with moderate-to-severe COPD. They also demonstrated a significant improvement in post-exercise oxygen saturation and a decrease in peak heart rate during the 6MWT with PEP, compared to controls without devices. However, in the same year, Wibmer et al.18 investigated the same kind of patients and found that, with a flow-dependent PEP in the range of 10–20cmH2O, patients walked 30.8m less than the controls without devices (352±92m with no PEP, 321±93m with PEP, P=.001). In their study, they demonstrated the efficacy of flow-dependent PEP on lung dynamic volumes, as total lung capacity, FRC and RV were significantly reduced after the 6MWT with PEP, compared to control group.

Although the three studies used different kinds of PEP, it seems that the application of lower pressures produces better results in terms of distance covered during the 6MWT. However, the “right” level of PEP during exercise is still largely understudied.

In this context, Martin et al.17 applied two levels of threshold PEP (13.8 and 2cmH2O) and demonstrated a significant decrease in dyspnea with the use of the higher level of PEP after a treadmill constant load test. However, these results cannot be compared with ours, because the authors applied PEP only after the conclusion of the exercise test, rather than during effort.

Recently, Bhatt et al.15 reported good results regarding exercise tolerance improvement in patients with severe COPD. They conducted a crossover trial investigating the effects of PLB on 6MWD and found that patients walked 34.9m more if they breathed with volitional PLB, compared to breathing normally (P=.002), with a concomitant reduction in RR after the test. Furthermore, they investigated the diaphragm excursion with ultrasonography and found a correlation between the improvements in 6MWD and the increase of diaphragm excursion with PLB. The authors explain their findings by suggesting that the PLB imposes a control and slows the RR, breaking the vicious circle of air trapping and dynamic pulmonary hyperinflation. This might translate into less neuro-ventilatory dissociation and a reduced perception of dyspnea. Low pressure PEP can be expected to act in a similar way.

As regards pulmonary function, low PEP has also been proposed for airway clearance in patients with COPD and hypersecretion. Venturelli et al.24 studied a mechanical device which applies 1cmH2O PEP at the mouth. They reported a significant improvement of IC at rest when patients breathe into the PEP device, compared to controls (+19.5% and +2.2%, P=.044). However, no data are available during exercise.

Monteiro16 and Padkao14 reported improvements in dynamic and static lung volumes, we cannot confirm the effect of PEP devices on post-exercise lung volumes. As we did not perform the spirometry immediately after the basal 6MWT, we do not have data on the behavior of the IC and the other lung volumes during the 6MWT without PEP. At present, we do not know if PEP produces any improvements in lung volumes compared to no PEP. We did not find any significant difference in the behavior of IC, SVC, ERV or FVC between PEP1 and PEP10. The application of low and high levels of PEP during the 6MWT proved to have the same effect on lung volumes.

Nevertheless, in our “ad hoc” analysis, we found that the application of low level PEP produces a greater improvement in the distance covered in the group of patients with no dynamic hyperinflation, compared to the group of patients who presented dynamic hyperinflation. This was in line with the findings of Callens et al., who observed that change in IC after walking does not correlate with 6MWD.25

In conclusion, it is difficult to compare the effects of PEP on exercise tolerance in patients with COPD, because the published studies are too heterogeneous, and use a variety of exercise protocols, types and levels of PEP and outcome measures. In order to clarify the effects of PEP on lung mechanics, it would be useful to evaluate the effect of PEP during a standardized exercise capacity test, such as the cardiopulmonary exercise test. As already discussed by Wibmer et al.,18 a better standardized evaluation of dynamic lung volumes, including at least some subsequent IC measures during the exercise, should also be carried out.

Future studies should lead to a more practical utilization of PEP devices. The effects of low level PEP during aerobic and strength training protocols, in the mid to long term, merit investigation.

In patients with severe COPD, the application of 1cmH2O PEP seems to improve the exercise tolerance to the same extent as 10cmH2O, with similar reported dyspnea. No effects were found on lung volumes after exercise. As previously reported protocols and levels and types of PEP vary widely, further studies are recommended to confirm our results and to clarify the mechanism of action on lung mechanics. Future studies should also investigate the effects of low levels of PEP on aerobic and strength training programs.