Healthy people

In healthy people, two main forms of overload have been imposed upon the respiratory muscles: (1) external loads at the mouth (intensity), and (2) voluntary hyperpnoea (increased breathing volume and flow rate) for extended periods (intensity and duration). In both cases, the training takes place daily, or at least three times per week (frequency).

Studies employing external loading have typically used load intensities in excess of 50% of inspiratory muscle strength (maximal inspiratory pressure: MIP), at a frequency of once or twice per day, for 5–7 days per week (McConnell & Romer, 2004). Loading at 50–70% of MIP typically yields task failure (see Ch. 6) within a duration of 30 breaths, or 2–3 minutes (intensity = 50–70%; duration = 30 breaths; frequency = twice daily). Statistically significant changes in muscle function have been measured within 3 weeks (Romer & McConnell, 2003), with a plateau in improvement occurring after around 6 weeks of training, despite continuous increases of the training load (Volianitis et al, 2001; Romer & McConnell, 2003). Changes in strength occurring within the first 2 weeks of strength training have traditionally been attributed to a neural adaptation process (Jones et al, 1989), i.e., improving the coordinated activation of synergistic muscles. Although this adaptation undoubtedly makes a contribution to the immediate short-term improvements seen in respiratory muscles, evidence from animal studies suggests that structural adaptation occurs within days of overload (Gea et al, 2000). Furthermore, in human beings, improvements in diaphragm thickness (8–12%) have been reported following just 4 weeks of inspiratory muscle training (IMT) (Downey et al, 2007) confirming the presence of rapid fibre hypertrophy in response to loading. These changes in diaphragm thickness were parallelled by improvements in maximal inspiratory pressure (MIP) (Downey et al, 2007). In placebo-controlled trials of IMT in healthy people, loads of 15% of MIP have been used as the placebo condition. When a 15% load is implemented with 30–60 repetitions it does not provide sufficient overload, as it fails to elicit changes in MIP (see (McConnell & Romer, 2004). Thus research suggests that inspiratory muscle overload in healthy people requires loads of 50–70% of MIP, eliciting muscle adaptations within 3–4 weeks.

Studies using hyperpnoea training have typically induced overload at intensities corresponding to 70% of maximum voluntary ventilation (MVV), for a duration of 15–40 minutes per day, at a frequency of once per day, for 4–5 days per week (Verges et al, 2008a). In the case of hyperpnoea training, overload is achieved by increasing the rate of air flow, with the inspiratory muscles working against the inherent resistance and elastance of the respiratory system (intensity = 70%; duration = 15–40 minutes; frequency = 4–5 days per week). Improvements in muscle function (endurance) are evident within 4 weeks (Verges et al, 2008a). There are currently no data to indicate the point at which functional improvements plateau after commencing of training.

The intensity and frequency dimensions of overload warrant specific mention, as they need to be balanced carefully so as not to tip the respiratory muscles into a state of ‘overtraining’. Most studies have implemented moderate-intensity IMT (50–70%) daily, but the intensity and frequency balance has yet to be studied systematically, or with consideration for other stimuli that overload the respiratory muscles. In the case of athletes this might be concurrent athletic training, and / or exposure to altitude, but in the case of patients it might be exercise training and / or the effects of an exacerbation. Preliminary unpublished data from healthy young soccer and rugby players (McConnell, unpublished observations) suggests that twice-daily high-intensity IMT (70–80% of MIP) may induce a state of chronic inspiratory muscle fatigue in athletes who are undergoing concurrent whole-body training. These training conditions appear to elicit suboptimal improvements in function. Thus, present evidence suggests that low-to-moderate-intensity loading (30–60% MIP) can be implemented daily, whereas high-intensity loading (>70% MIP) should be implemented no more than once every other day.

Patients

In the majority of reported studies in patients with respiratory disease, training has been undertaken by imposing an external load at the mouth (intensity) for 10–30 minutes (duration). Typically, training has been undertaken for 2 to 3 months, but structural and biochemical adaptations to the inspiratory muscles are evident within 6 weeks (Ramirez-Sarmiento et al, 2002). A study in patients with chronic obstructive pulmonary disease (COPD) has demonstrated the time course of changes in strength over a 12-month intervention (Weiner et al, 2004). Weiner and colleagues (Weiner et al, 2004) noted the largest improvement in MIP during the first 3 months of their study (32%), followed by smaller increases (~ 6%) for the four subsequent 3-month blocks of IMT. Training sessions have typically been conducted in continuous bouts lasting 10–30 minutes, 1–2 times a day, for 5–7 days per week.

The intensity dimension has not been studied extensively in patients, but data from seven studies of patients with COPD collated by Pardy & Rochester (1992) suggested a significant positive relationship between the percentage increase in MIP and the relative magnitude of the inspiratory training load. In other words, the higher the load relative to the subject’s MIP, the greater was the increase in MIP induced by training. The collated data suggest that, to achieve a 20% increase in MIP, a load of at least 30% of maximum strength is required. This suggestion is also supported by data from a 2002 meta-analysis of IMT (Lotters et al, 2002), and in a study that compared the efficacy of high- (52% MIP) and low-intensity (22%) IMT (Preusser et al, 1994); high-intensity IMT increased MIP by 35% (p < 0.05), whereas low-intensity IMT increased MIP by only 10% (p > 0.05) (Preusser et al, 1994). Collectively, the literature supports the need for training loads to exceed 30% MIP, but the question of whether more is better when it comes to the magnitude of inspiratory loading has not been examined systematically. A handful of studies have reported the effects of high-intensity training in patients with COPD, and these suggest that when loads are 68% of MIP (Sturdy et al, 2003), ‘the highest tolerable inspiratory threshold load’ (Hill et al, 2006) or ‘80% of maximal effort’ (Enright et al, 2006), then greater increases in MIP are achieved (29–41%) compared with low-to-moderate intensities (15–23%; Geddes et al, 2008).

However, it is important to note that in all of the studies employing high-intensity IMT the frequency of training has been only 3 days per week, compared with twice daily IMT in the studies using low to moderate loads (i.e., three sessions per week compared with 14). This is potentially very important from a practical point of view, as it suggests that high-intensity training may be far more time efficient, as well as more effective. However, a note of caution should also be expressed, as daily high-intensity IMT may overload the muscles to the extent that chronic inspiratory muscle fatigue and suboptimal adaptations are elicited (see above). Practical suggestions regarding load setting can be found in Chapter 6.

Studies of IMT in patients have typically been of much longer duration than those in healthy young athletes (3 months vs 4–6 weeks). As mentioned previously, Weiner and colleagues noted the largest improvement in MIP during the first 3 months of their study, followed by a gradual plateau of improvement (Weiner et al, 2004). Similar observations were made by Larson et al (1988) after 1 month of training, and by Lisboa et al (1997). This ‘plateau’ effect is also apparent in studies on healthy people, where MIP increases most rapidly during the first 3 weeks, then rising more slowly to a plateau by around 6 weeks (Volianitis et al, 2001; Romer & McConnell, 2003). The development of a plateau cannot be ascribed to a lack of load progression (increasing the training load to accommodate increases in MIP) as it occurs regardless of this measure. Instead, it is a reflection of a basic property of muscle adaptation to strength-training stimuli (Moritani & deVries, 1979; Hakkinen et al, 1987), which necessitates periodic changes in the nature of the training stimulus in order to maintain the adaptation process; this is one of the reasons why athletes periodize their training.

Training for strength

Generally, respiratory muscles respond to high-load–low-frequency loading with a strength-training response (Pardy & Rochester, 1992; Tzelepis et al, 1994a; Romer & McConnell, 2003). However, as well as load specificity, there is also an element of flow specificity that must be borne in mind as the two are interrelated (Tzelepis et al, 1994a; Romer & McConnell, 2003). This is because of the limitations imposed by the force–velocity relationship of muscles (see Ch. 1); high loads cannot be overcome at high velocities of muscle shortening. Training stimuli with high loads and low velocities (e.g., a Mueller manoeuvre) elicit increases in MIP, but do not elicit increases in maximal shortening velocity (peak inspiratory flow rate) (see Ch. 4, Fig. 4.2). Conversely, training with low loads and high velocities of shortening (e.g., unloaded hyperpnoea) elicit increases in maximal shortening velocity, but not MIP (see Ch. 4, Fig. 4.2) (Tzelepis et al, 1994a; Romer & McConnell, 2003). Interestingly, training stimuli with intermediate loads and shortening velocities elicit improvements in both qualities (Tzelepis et al, 1994a; Romer & McConnell, 2003), which arguably provides the ‘best of both worlds’ (see Ch. 4, Fig. 4.2).

A number of studies have now demonstrated in healthy people (Enright et al, 2006; Downey et al, 2007) and patients (Ramirez-Sarmiento et al, 2002; Enright et al, 2004; Chiappa et al, 2008; West et al, 2009) that the increase in MIP that follows strength training of the inspiratory muscles is secondary to hypertrophy.

Training for endurance

An endurance-conditioning response can be elicited with prolonged low-load–high-frequency contractions, which have typically been imposed upon the respiratory muscles using prolonged voluntary hyperpnoea (Boutellier & Piwko, 1992), but endurance can also be improved through strength training (Belman & Shadmehr, 1988; Harver et al, 1989). There is a common misconception that muscle endurance can be improved only using a specific endurance-training stimulus. However, stronger muscles perform a given task at a lower percentage of their maximum capacity than weaker muscles, which has beneficial consequences for fatigue resistance (endurance) (Belman & Shadmehr, 1988). Thus, inspiratory muscle strength training provides a ‘dual-conditioning’ response. There is no evidence that a specific endurance-training stimulus, such as hyperpnoea, improves MIP (Leith & Bradley, 1976; O’Kroy & Coast, 1993); indeed, this would not be expected as strength improves only when the tension within muscles is increased by the imposition of an external load. Collectively, the data suggest that training regimens with a moderate strength bias have the capacity to improve maximal strength, velocity of shortening and power output (Romer & McConnell, 2003) as well as endurance (Romer & McConnell, unpublished observations). This versatility supports the implementation of training with a bias towards strength at a moderate intensity.

The effect of lung volume (muscle length)

To date, only one study has examined whether the lung volume at which IMT occurs has any influence upon training outcomes (Tzelepis et al, 1994b). The data indicate that improvements in inspiratory muscle strength are specific to the lung volume at which training occurs (Tzelepis et al, 1994b). When three groups of healthy participants performed 6 weeks of repeated static maximum inspiratory manoeuvres at one of three lung volumes (residual volume, functional residual capacity (FRC), or FRC plus one-half of inspiratory capacity), the greatest improvements in strength occurred at the volume at which the participants trained. In addition, the improvements were significantly greater for those who trained at low lung volumes. Furthermore, the range of lung volume over which strength was increased was also greatest for those who trained at low lung volumes. These data suggest that IMT should be conducted over the greatest range of lung volumes possible, commencing as close as possible to residual volume.

A caveat that must be borne in mind in the context of lung volume specificity is the volume–pressure relationship of the inspiratory muscles (see Ch. 1). The inspiratory muscles become progressively weaker as the lungs inflate, such that under conditions of inspiratory loading the breath may be curtailed before the lungs are completely full. This ‘clipping’ of inspired volume is influenced by: (1) the magnitude of the load (occurring earlier with heavier loads), and (2) the fatigue state of the inspiratory muscles (occurring earlier in the presence of fatigue) (see also Ch. 6, Fig. 6.2). This means that, for most resistance IMT devices, a compromise must be struck between the magnitude of the load and the volume of the breath, as it is impossible to achieve high loads and high volumes simultaneously. The impact of these issues upon training load selection is considered in Chapter 6. The influence upon the design of IMT products will be considered at the end of this chapter.

Detraining

Unfortunately, the extent and time course of inspiratory muscle detraining are not well documented, but two studies of resistance IMT do shed some light on these issues. In healthy young adults, Romer & McConnell (2003) documented regression of IMT-induced changes in inspiratory muscle function (9 weeks of three differing IMT regimens) over an 18-week period of detraining. Decrements were observed at 9 weeks, with no further changes in strength-related measures at 18 weeks post-IMT. In contrast, endurance continued to decline between 9 and 18 weeks of detraining (Romer & McConnell, unpublished observations). Inspiratory muscle function remained significantly above baseline at 18 weeks, with a loss of 32% of the improvement in strength, 65% of the improvement in maximum shortening velocity and 75% of the improvement in inspiratory muscle endurance.

In patients with COPD, Weiner et al (2004) observed the detraining response of a group of COPD patients who had completed a 3-month intensive IMT programme. The detraining group undertook sham training (inspiratory load of 7 cmH₂O) for the next 12 months and were reassessed at 3-month intervals. After 3 months of detraining, both MIP and inspiratory muscle endurance remained elevated compared with baseline (MIP 19%, endurance 22%), but after 12 months they were not significantly different from baseline. Collectively, these data suggest that inspiratory muscles respond in a similar manner to other muscles when a training stimulus is removed (Mujika & Padilla, 2000a, b) and that most of the losses of function occur within 2 to 3 months of the cessation of training.

Maintenance

On a more positive note, the two detraining studies described above (Romer & McConnell, 2003; Weiner et al, 2004) also demonstrated that IMT-induced improvements in inspiratory muscle function can be sustained with maintenance training programmes in which training frequency is reduced. Training frequency can be reduced by as much as two-thirds without loss of function, i.e., to 2 days per week in healthy adults (Romer & McConnell, 2003) and to 3 days per week in patients with COPD (Weiner et al, 2004).

In summary, the literature supports the notion that the general training principles of overload, specificity and reversibility apply as much to the training of respiratory muscles as they do to limb muscles. This means that respiratory training interventions should apply these principles in order to obtain specific functional outcomes (see Ch. 6).