This was a randomized, placebo-controlled trial conducted in accordance with World Medical Association guidelines for research on humans. Our institutional ethics board approved all protocols, and all the patients gave informed consent before participating in the study. Fourteen community-based patients with COPD (males, 66 ± 5 years) were selected for the study from a hospital respiratory clinic. The COPD diagnosis was determined from a clinical history consistent with chronic bronchitis and/or emphysema, a long history of cigarette smoking, and pulmonary function tests revealing fixed airflow obstruction (FEV₁ < 75% predicted, and FEV₁/FVC ratio < 65%) (13). Subjects were sedentary and were observed during a 4-week period while their regular treatment was maintained to evaluate functional status and to verify clinical stability. Patients showing severe hypoxemia (i.e., Pa_O2 lower than 60 mm Hg breathing room air); asthma; coronary disease; undernourishment (body mass index < 20 kg/m²); chronic metabolic diseases (e.g., diabetes, hypo- or hyperthyroidism); orthopedic diseases; previous abdominal or thoracic surgery; or treatment with steroids, hormones, or cancer chemotherapy were not considered eligible for the study. The patient's characteristics are summarized in Table 1

Nutritional assessment, pulmonary function tests, inspiratory and expiratory muscle strength, and transdiaphragmatic pressure (14) during both quiet breathing and maximal sniff maneuvers were measured in each patient and compared with reference values (15–19), as described elsewhere (20). The inspiratory muscle pressure–time index (PT_I) while the patient was seated during tidal breathing was calculated according to the equation PT_I = (Pes/Pes_max) · (T_I/T_TOT), where Pes is the esophageal pressure, Pes_max is the esophageal pressure during sniff maneuvers, and T_I/T_TOT is the duty cycle.

Exercise capacity was evaluated using two tests. First, a submaximal exercise test was performed to assess the distance the patient was able to walk in 6 minutes along a measured flat corridor as described by Butland and coworkers (21). Second, maximal exercise capacity was evaluated during an incremental exercise test to volitional exhaustion on a cyclergometer referenced to values from Jones (22)

Specific inspiratory muscle endurance was assessed during two different threshold inspiratory loading tests performed using a device that was similar to the one described by Nickerson and Keens (23). In the first loading test, the volunteers breathed against incremental loads (∼ 8 cm H₂O every 2 minutes) until maximal sustainable threshold pressure was reached (Pth_max) (24). In the second loading test, subjects breathed against a submaximal constant load (equivalent to 80% of maximal threshold pressure) until exhaustion. The elapsed time was defined as the inspiratory sustainable threshold endurance time (Tth₈₀, in minutes).

Biopsies from the external intercostal (inspiratory muscle) and vastus lateralis (control) muscles were taken before and after the inspiratory muscle training period (25). The pretraining biopsies were taken from the nondominant or dominant side, as previously randomized. Biopsies from the external intercostal muscles were taken along the anterior axillary line at the sixth intercostal space. Biopsies from the vastus lateralis were taken from the middle portion of the thigh. The post-training biopsies were obtained from the same anatomical site but from the contralateral side. The size of the muscle samples was approximately 5 × 5 × 5 mm. Each biopsy was quick-frozen in isopentane cooled in liquid nitrogen and stored at −70°C. Ten-micrometer thick sections were cut, varying the inclination of the holder by 5° increments until the minimum cross-sectional area was obtained, which was defined as truly transverse (26, 27). Consecutive cross-sections were processed by immunohistochemical techniques using monoclonal antibodies directed against myosin heavy chain (MyHC) isoforms type I and type II (Figure 1)

Figure 1. Microphotographs of consecutive cross-sections of external intercostal muscle processed by conventional hematoxilin (A) and immunohistochemical techniques. For the latter, monoclonal antibodies directed against MyHC isoform type I and type II were used (B and C, respectively). For details, see text.

Download Figure | Download Powerpoint

[More] [Minimize]

The subjects received either inspiratory muscle training or sham training for 30 minutes, breathing through a threshold inspiratory device 5 days a week for 5 consecutive weeks. Appropriate personnel supervised training sessions. Loaded breathing was intermittent for 3-minute periods, with a 2-minute rest period in between. The patients breathed against a load equivalent to 60% of their maximal sustainable inspiratory pressure (which represented approximately 40–50% of the initial PI_max). The load could be increased depending on patient tolerance. Patients who received sham training breathed through the same inspiratory muscle training device with no additional load.

Values throughout the text and tables are expressed as mean ± SD. Baseline and post-training data were compared within groups using the nonparametric (Wilcoxon) tests for paired samples. The Mann–Whitney U-test was used to compare data between groups. An α (p) value less than or equal to 0.05 was considered statistically significant.

Sixteen patients were initially recruited for the study. Two of them were excluded, one due to pulmonary infection at the beginning of the training period and the other due to noncompliance with the training dates. Complications derived from either the functional evaluations or muscle sampling were not detected in any patient. The general characteristics of the study population are shown in Table 1. Changes in conventional pulmonary function tests (Table 2)

Figure 2. Main changes in both structural and functional variables expressed as the difference (in absolute values) between initial versus final values. Dark bars represent the values from inspiratory muscle training group, whereas the light bars represent the values from the sham training group. *p Value less than 0.05 when compared with the intragroup initial value; n.s. refers to nonsignificant difference.

Download Figure | Download Powerpoint

[More] [Minimize]

The overall function of the diaphragm and accessory inspiratory muscles showed significant changes after training in the inspiratory muscle training group (Figure 2) as expressed by an increase, first, in the inspiratory muscle strength (PI_max, Pes_max, and Pdi_max), and second, in the inspiratory muscle endurance (Table 3). Neither the breathing pattern nor the strength of the expiratory muscles (PE_max) showed differences after training (Table 3).

Before training, the size of the external intercostal fibers was found to be 47 ± 8 μm, and a total of 46 ± 18% of the fibers expressed type I MyHC. The inspiratory muscle training clearly associated with structural changes in the muscle, as assessed by both fiber type distributions and fiber size. Specifically, both the proportion of type I fibers (p < 0.05) and the size of type II fibers (p < 0.05) increased after training (Table 4

Figure 3. Microphotograph showing immunohistochemical staining using anti–MyHC-I monoclonal antibody for fiber typing and morphometric analyses in cross-sectional sections from the external intercostal muscles of a patient who received specific inspiratory muscle training. In this figure, dark fibers correspond to fibers expressing MyHC type I, whereas light fibers indicate fibers expressing another MyHC phenotype. For details, see text. Pre and post refer to pre- and post-training biopsy, respectively.

Download Figure | Download Powerpoint

[More] [Minimize]

Before training, the mean size of vastus lateralis fibers was found to be 57 ± 10 μm. A total of 30 ± 11% of the fibers expressed type I MyHC. The inspiratory muscle training did not promote structural changes within the vastus lateralis, as assessed by fiber size or fiber type distributions (see Table E1 in the online data supplement).

The present study was aimed at evaluating the effects of a specific inspiratory muscle training on the structure of inspiratory muscles in community-based patients with COPD. This could be controversial regarding the current state of respiratory muscle adaptations. In fact, the diaphragm of nontrained patients with COPD exhibit structural changes that presumably represent “adaptive effects.” Studies of Levine and coworkers (30), Mercadier and coworkers (31), and one study of our group (32) demonstrate that the diaphragm shows an increase of type I fibers, MyHC-I, and mitochondria (supporting a fiber type transformation), whereas the length of the sarcomere decreases (supporting adaptation to chronic diaphragm flattening) (32). A reasonable presumption is that other primary respiratory muscles (such as the external intercostals) could exhibit similar adaptations. From a clinical point of view, however, such structural adaptations only appear to partially restore their functional capacity. In fact, inspiratory muscles of patients with COPD show a lower capacity to generate maximal pressures (strength) or tolerate submaximal inspiratory loads (endurance) (4). This muscle dysfunction could be explained by the coexistence of other factors (33–35) with deleterious effects such as (1) geometric changes of the thorax due to increased lung volume and shortening of the diaphragm fibers; (2) intrinsic changes within the muscles due to malnutrition, ionic, or arterial gas disorders; or (3) the effect of drugs or concomitant diseases or conditions (e.g., aging). Altogether, these arguments have prompted several investigators to hypothesize that inspiratory muscle training could offer clinical benefits to patients with severe COPD. However, the results of previous studies have been controversial.

The size of external intercostal fibers was found to be within the normal range before specific training (48). The most important finding of the present study is that inspiratory muscle training induced structural changes within the trained muscles of stable patients with COPD. A few experimental studies have shown that the diaphragm and other inspiratory muscles show structural changes in animals submitted to inspiratory loads (11). However, the authors have been unable to find any study evaluating structural changes in human inspiratory muscles after specific training.

The findings of the present study highlight three main concepts. First, muscle response; the study demonstrates that external intercostal muscles of patients with COPD retain the capacity to be remodeled (conditioned) after a short-term loading period. The muscles exhibited a classical response to training that would be predictable in limb muscles. Similar findings have been described in the peripheral muscles of patients with COPD after general muscle training (49). The novelty of the present report is that this response is demonstrated in an inspiratory muscle group from patients with severe airflow obstruction. Second, functional and structural changes; the results allow us to hypothesize that the increase in inspiratory muscle endurance and strength after specific training could be related to the observed switching of MyHC isoforms (as assessed by the increase in fibers expressing MyHC-I) and to the increase in fiber size (mainly in type II fibers). Other factors such as perceptual adaptation to additional inspiratory loading (e.g., dyspnea desensitization), learning of specific maneuvers, or even a placebo effect could also have participated in improving inspiratory muscle strength and endurance. And third, specificity; we found that inspiratory muscle training had specific functional and structural effects only on the trained (inspiratory) muscles. The study included data from an unaffected (limb) muscle as a negative control. No changes were observed either in the fiber size or in fiber type proportions of the vastus lateralis (control muscle) when results before and after training were compared. Similarly, a transfer effect to other respiratory muscle groups (e.g., changes in function of the expiratory muscles) was not found. These facts support the conclusion that inspiratory training has a specific effect only on the trained muscles and allow us to hypothesize that structural adaptation occurred only in the inspiratory muscles. However, the design of the study does not permit total confidence in the structural changes within the external intercostals are representative of inspiratory muscles in general. The diaphragm is the most important of the inspiratory muscles, but there are obvious ethical and practical difficulties in repeatedly accessing the diaphragm either from healthy subjects or from stable patients with COPD, even if thoracotomy was performed for other reasons (e.g., lung volume reduction, lung cancer, or transplant).

The principal limitations of the present study are the relatively small number of subjects studied and the possibility that the results could be related to regional differences in the external intercostal muscles within a given volunteer. The authors believed that a greater number of volunteers was not essential because the results reached statistical significance from both structural and functional points of view, and the study that used relatively high invasive procedures. The results do not appear to be related to regional differences within the external intercostal muscles. A previous study demonstrated that biopsies from the external intercostal muscles taken from different anatomic regions are comparable in patients with COPD, even when the samples were taken either from the dominant or from the nondominant side (50). In addition, we believe that anatomical knowledge and experience allow us to ensure that muscle samples came from the external intercostals in all cases. The experience is the result of both anatomical reviews and practical evidence. At least nine arguments should be highlighted: (1) the technique is an “open” biopsy, in contrast to “blind” needle techniques; (2) we have reviewed and we have enough knowledge of the specific body zone; (3) we have practiced in cadaver models; (4) we normalized the body location for the muscle biopsies; (5) we performed a sequential and careful dissection technique with identification of all tissue plans (fascia, muscle); (6) we discarded a potential error by taking biopsies from the pectoralis because the “anatomical window” to access the external intercostals is below and relatively far from the lower insertions of the pectoralis muscles; (7) the biopsy was taken from the intercostal space; (8) we identified the orientation of the muscle fibers that is clearly different for external and internal intercostals; and (9) due to intimate contact between internal intercostals and the pleura, the absence of yatrogenic pneumo-thorax confirms that such muscles were not sampled.

This is the first report demonstrating that structural changes occur in respiratory muscles of patients with COPD after specific training, in association with improvement in the inspiratory muscle function. The authors believe that this evidence offers additional information, improves the knowledge of the physiologic basis of inspiratory muscle training, and reinforces the inclusion of inspiratory muscle training as a part of pulmonary rehabilitation. The results are innovative and therefore could not have been directly extrapolated either from previous animal studies (because many factors present in COPD do not exist in animal models) or from previous studies assessing the effects of peripheral muscle training (because pretraining phenotype is completely different from that reported for the respiratory muscles). Our selection criteria excluded patients showing malnutrition, respiratory failure, hypercapnia, or systemic steroid treatment, which are features usually found in patients with COPD. However, these conditions would introduce confounding factors to the study. Further studies evaluating the potential effect of inspiratory muscle training in patients with these associated conditions, the time course of the muscle remodeling process, and the minimal external loading capable of muscle adaptation seem to be warranted.

The present study demonstrates that the external intercostal muscles of patients with COPD have the capacity to express structural remodeling after specific inspiratory training. Both the proportion of type I fibers and the size of type II fibers were found to increase after training. These structural adaptations could explain, in part, the functional improvements observed in the trained muscles (increased inspiratory muscle strength and endurance) after specific training.