*2.2.2. Lung ultrasound*

Lung ultrasound (LU) is a simple, non-invasive and radiation-free methodology [104] mainly used in critical care, emergency medicine, trauma surgery and pulmonary medicine for diagnostic purposes [105]. However, due to its practical and secure character, which enables its use as often as required, LU has become an attractive alternative imaging technique for monitoring patients on whom thoracic computed tomography (CT) cannot be performed on a routine basis or where chest X-ray presents serious limitations in terms of sensitivity and specificity [106]. In fact, the international evidence-based guidelines for LU recommend this technique to monitor aeration changes and the effects of therapy in a number of acute respi‐ ratory diseases, including acute pulmonary edema, acute respiratory distress syndrome, acute lung injury, community-acquired pneumonia, ventilator-associated pneumonia and recovery from lavage of alveolar proteinosis (level A) [105].

Usually, LU is performed using a 3–5 MHz convex transducer to visualise deeper lung structures [107, 108]. A high-frequency 5–12 MHz linear probe is most effective in visualising the chest wall, pleura and the lung peripheral parenchyma [108]. A complete examination of the chest requires longitudinal, transversal and oblique-array probes to be placed along the rib spaces, proceeding from top to bottom in the ventral-dorsal direction, along of 12 regions of interest (parasternal, medial clavicular, anterior axillary, medial, and posterior right and left chest walls) [106, 108]. Anterior examination should be performed with the patient in the supine or semi-lateral position [106] while posterior examination should take place with the patient seated [107].

LU has shown to be reliable in the diagnosis of several acute respiratory conditions, such as pneumothorax (sensitivity of 65–100%; specificity of 78–100%) [108-110], including the diagnosis of this condition in patients with CF (specificity of 100%) [109], interstitial syndrome (sensitivity and specificity of 94%), lung consolidation (sensitivity of 90 to 95%; specificity of 95%), pleural effusion (sensitivity of 90 to 100%; specificity of 100%) [110] and atelectasis (sensitivity of 88%; specificity of 89%) [104]. One study assessed the inter-subject reliability of LU in the detection of atelectasis and reported a very good agreement (kappa=0.90; 95%CI 0.75 to 1) [104]. When compared with other imaging equipment (e.g. chest X-ray and MRI), this measure showed similar or even better reliability results [104, 107, 110].

No data has been found regarding the use of LU to monitor respiratory physiotherapy interventions in stable or exacerbated patients with CF. Nevertheless, considering its high accuracy in detecting signals commonly presented in CF exacerbations, such as atelectasis and consolidation [111], and its increasing impact in the management of acute respiratory condi‐ tions [105], it is reasonable to conclude that LU may be a promising measure to monitor the effectiveness of respiratory physiotherapy in patients with acute exacerbations of CF. Consid‐ ering stable CF, one study used LU to characterise diaphragm thickness as a way to infer about its muscle mass. The authors reported an excellent inter- and intra-subject agreement (90% and 91%, respectively) and showed that LU was capable of detecting differences between patients with low and high rates of fat-free mass [112]. From these findings, it can be hypothesised that LU may also play a role in the assessment of the effectiveness of inspiratory muscle training and general exercise training programmes in increasing the diaphragm thickness of patients with stable CF, and thus, its muscle mass. Nevertheless, studies assessing the LU validity, reliability and responsiveness to change are needed before it can be recommended as an outcome measure for respiratory physiotherapy interventions.

At this point, there is no evidence to recommend LU as an outcome measure for respiratory physiotherapy in patients with CF. However, the advantages presented by LU over other imaging measures and its good performance as a diagnostic tool, should motivate further investigation on the validity and reliability of this measure to assess respiratory physiotherapy interventions in this population.

### *2.2.3. Fat-free mass*

Despite the scarce evidence of computerised respiratory sounds in CF, it seems that this measure might offer potential to assess the short- and long-term effects of respiratory physi‐ otherapy interventions in these patients. However, more research is still required to determine the parameters of computerised respiratory sounds (i.e., number, frequency, duration) that are more sensitive to change after an intervention and to establish the reference values that will allow physiotherapists to interpret with confidence the results obtained from the computerised

Lung ultrasound (LU) is a simple, non-invasive and radiation-free methodology [104] mainly used in critical care, emergency medicine, trauma surgery and pulmonary medicine for diagnostic purposes [105]. However, due to its practical and secure character, which enables its use as often as required, LU has become an attractive alternative imaging technique for monitoring patients on whom thoracic computed tomography (CT) cannot be performed on a routine basis or where chest X-ray presents serious limitations in terms of sensitivity and specificity [106]. In fact, the international evidence-based guidelines for LU recommend this technique to monitor aeration changes and the effects of therapy in a number of acute respi‐ ratory diseases, including acute pulmonary edema, acute respiratory distress syndrome, acute lung injury, community-acquired pneumonia, ventilator-associated pneumonia and recovery

Usually, LU is performed using a 3–5 MHz convex transducer to visualise deeper lung structures [107, 108]. A high-frequency 5–12 MHz linear probe is most effective in visualising the chest wall, pleura and the lung peripheral parenchyma [108]. A complete examination of the chest requires longitudinal, transversal and oblique-array probes to be placed along the rib spaces, proceeding from top to bottom in the ventral-dorsal direction, along of 12 regions of interest (parasternal, medial clavicular, anterior axillary, medial, and posterior right and left chest walls) [106, 108]. Anterior examination should be performed with the patient in the supine or semi-lateral position [106] while posterior examination should take place with the

LU has shown to be reliable in the diagnosis of several acute respiratory conditions, such as pneumothorax (sensitivity of 65–100%; specificity of 78–100%) [108-110], including the diagnosis of this condition in patients with CF (specificity of 100%) [109], interstitial syndrome (sensitivity and specificity of 94%), lung consolidation (sensitivity of 90 to 95%; specificity of 95%), pleural effusion (sensitivity of 90 to 100%; specificity of 100%) [110] and atelectasis (sensitivity of 88%; specificity of 89%) [104]. One study assessed the inter-subject reliability of LU in the detection of atelectasis and reported a very good agreement (kappa=0.90; 95%CI 0.75 to 1) [104]. When compared with other imaging equipment (e.g. chest X-ray and MRI), this

No data has been found regarding the use of LU to monitor respiratory physiotherapy interventions in stable or exacerbated patients with CF. Nevertheless, considering its high accuracy in detecting signals commonly presented in CF exacerbations, such as atelectasis and consolidation [111], and its increasing impact in the management of acute respiratory condi‐

measure showed similar or even better reliability results [104, 107, 110].

auscultation.

*2.2.2. Lung ultrasound*

50 Cystic Fibrosis in the Light of New Research

patient seated [107].

from lavage of alveolar proteinosis (level A) [105].

Fat free mass (FFM) is the component of body mass that represents muscle mass and protein stores [113] and it is a critical determinant of maximal exercise capacity [114]. It is known that, in patients with CF, lower FFM is also associated with lower FEV1 percentage predicted and more frequent respiratory exacerbations [115, 116]. Maintaining appropriate levels of FFM is therefore crucial for maintaining the overall functional capacity in patients with CF [114].

The gold standard to assess FFM are the 4-C models, which divide body weight into fat, water and remaining fat-free dry tissue, with the last item further divided into proteins and minerals. The 4-C models require measurements of body weight, body water, body volume and bone mineral [117] and thus several specialised equipment, such as dual-energy X-ray, measures of air displacement plethysmography (BOD POD) and D2O analysis, are needed. The partial measures are then pulled together in a predictive equation [118]. Although all devices involved are valid and reliable (intraclass correlation coefficient>0.99), the improved accuracy of the 4- C models may be offset by the potential propagation of errors due to the inherent measurement error of each device used to assess each variable [118]. Also, it appears not to be suitable for widespread clinical use due to the need for specialised equipment, experienced personnel and large amount of time. Thus, its main value lies in the quality of its evidence in supporting treatment approaches, rather than in routine practical application [117, 119].

For application of respiratory physiotherapy clinical practice, 2-C models, which require only the use of one device, are the most recommended and have shown large correlations with the 4-C models. These are bioelectrical impedance (r>0.79, R2 =0.70), CT scans (r>0.83, R2 =0.96), MRI (r>0.91), and specially dual-energy X-ray (r>0.91) [120, 121]. Indirect strategies involving calculations based on anthropometric measurements, such as skinfold thickness analysis, are controversial. Measuring FFM by measurement of skinfolds implies that no index of this component of weight is directly measured, and thus it is based on the assumption that measuring fatness reflects lean body mass as well. Therefore, its use has not been recommend‐ ed [117]. Nevertheless, studies that show medium correlations between measurements of skinfold thickness and 4-C models (R2 =0.62) exist [120] and support the usability of this measure to monitor FFM irrespective of the clinical severity of CF [122].

FFM has been used as an outcome measure in respiratory physiotherapy interventions for children and adults with CF, mainly for the assessment of different types of exercise training. In the study of Selvadurai et al. (2002), 66 children with exacerbated CF significantly increased their FFM after five sessions of exercise training (aerobic or resistance training) or chest physiotherapy, independently of the protocol applied (Cohen's dz: 1.65 to 5.22) [123]. Sosa et al. (2012) did not find improvements in the percentage of FFM of children with stable CF following an 8-week exercise training protocol (78.1 to 79.4%; Cohen's dz 0.47) [119], however, significant differences were reached when a component of inspiratory muscle training was added (81.6 to 82.6%; Cohen's dz 0.85) [124]. Gruber et al. (2014) compared the effect of a 6 week interval exercise training vs. standard exercise training in 43 adults with CF [125]. Significant improvements in FFM were only observed in the standard exercise training group (41.7 to 43.3 kg; Cohen's dz 0.29). These four studies, with its medium and large effect sizes, demonstrated that FFM is an adequate outcome measure to assess the effectiveness of respiratory physiotherapy or just exercise training protocols in patients with CF.

Despite the limited evidence available, FFM assessed with 2-C or 4-C models appears to be a valid, reliable and usable measure to assess respiratory physiotherapy interventions in CF.

### *2.2.4. Inspiratory muscle strength*

One of the functions of the muscles is to develop strength. In the specific case of the inspiratory muscles, strength is usually estimated as pressure [126]. Several techniques have been described to measure inspiratory muscle strength, however, the maximum static inspiratory pressure at the mouth (Pimax) is one of the most widely used [126].

The assessment of Pimax requires patients to make a maximum inspiratory (Mueller manoeu‐ vre) effort at or near the residual volume, maintained for at least 1–1.5 seconds. These tests are volitional and require patient's full cooperation. For these reasons, this technique is usually performed by an experienced health professional to assure adequate instruction and encour‐ agement. The manoeuver is repeated until the variation between measures is less than 20%. Pimax is usually expressed in absolute values (cmH2O) or as a percentage of the predicted values.

Pimax is a simple, low-cost and well tolerated technique. In addition, through the hand-held and portable electronic pressure transducers, the technique is easily used by physiotherapists in a wide range of clinical settings (e.g., primary care, hospital wards, intensive care units). Another advantage of using this technique is the availability of reference values for healthy children, adults and the elderly, enabling the interpretation of results [127, 128].

The reliability of this measure is well established in people with non-CF bronchiectasis [129] and healthy people [130]. In patients with CF, only one study was found assessing the test– retest reliability of Pimax [131]. Pimax had an excellent coefficient of reliability (89%) and intraclass correlation coefficient (0.88) in patients with CF (n=20), in line with the results obtained in healthy individuals (coefficient of reliability 91%; intraclass correlation coefficient 0.87) [131]. One of the disadvantages of this study is the fact that it was conducted only with adult patients with CF (22.7±3.4 years old). The reliability of Pimax in children with CF is still unexplored.

Due to its simplicity and adequate reliability, Pimax has been used as an outcome measure of inspiratory muscle training programmes. In the study by de Jong et al. (2001), seven patients with CF improved their Pimax from 105 to 123cmH2O after 6 weeks of inspiratory muscle training (Cohen's dz 0.82) [132]. Enright et al. (2004) investigated the effect of an 8-week inspiratory muscle training with a high (n=9) and low (n=10) training intensity. Large effects were found using the Pimax, with improvements from 114–134cmH2O to 155–159cmH2O (Cohen's dz 1.01 and 1.34) [43]. Amelina et al. (2006), after a 6-week inspiratory muscle training, found that patients with CF (n=10) had a mean improvement from 77% to 91% of the predicted Pimax (p=0.023; Cohen's dz 0.73) [133]. These three studies, with its medium and large effect sizes, demonstrated that Pimax is an adequate outcome measure to assess the effectiveness of inspiratory muscle training.

Pimax has also been used as an outcome measure to determine the effectiveness of noninvasive ventilation during chest physiotherapy [39]. It was found that Pimax was maintained following ACBT assisted with non-invasive ventilation, resulting in a significant difference compared with ACBT alone (mean difference from standard treatment 9.04cmH2O, 95%CI 4.25 to 13.83, p=0.006) [39].

Despite the limited evidence available, it seems that physiotherapists can confidently rely on Pimax to assess the effectiveness of respiratory physiotherapy interventions in patients with CF.

### *2.2.5. Inspiratory muscle endurance*

measuring fatness reflects lean body mass as well. Therefore, its use has not been recommend‐ ed [117]. Nevertheless, studies that show medium correlations between measurements of

FFM has been used as an outcome measure in respiratory physiotherapy interventions for children and adults with CF, mainly for the assessment of different types of exercise training. In the study of Selvadurai et al. (2002), 66 children with exacerbated CF significantly increased their FFM after five sessions of exercise training (aerobic or resistance training) or chest physiotherapy, independently of the protocol applied (Cohen's dz: 1.65 to 5.22) [123]. Sosa et al. (2012) did not find improvements in the percentage of FFM of children with stable CF following an 8-week exercise training protocol (78.1 to 79.4%; Cohen's dz 0.47) [119], however, significant differences were reached when a component of inspiratory muscle training was added (81.6 to 82.6%; Cohen's dz 0.85) [124]. Gruber et al. (2014) compared the effect of a 6 week interval exercise training vs. standard exercise training in 43 adults with CF [125]. Significant improvements in FFM were only observed in the standard exercise training group (41.7 to 43.3 kg; Cohen's dz 0.29). These four studies, with its medium and large effect sizes, demonstrated that FFM is an adequate outcome measure to assess the effectiveness of

measure to monitor FFM irrespective of the clinical severity of CF [122].

respiratory physiotherapy or just exercise training protocols in patients with CF.

pressure at the mouth (Pimax) is one of the most widely used [126].

Despite the limited evidence available, FFM assessed with 2-C or 4-C models appears to be a valid, reliable and usable measure to assess respiratory physiotherapy interventions in CF.

One of the functions of the muscles is to develop strength. In the specific case of the inspiratory muscles, strength is usually estimated as pressure [126]. Several techniques have been described to measure inspiratory muscle strength, however, the maximum static inspiratory

The assessment of Pimax requires patients to make a maximum inspiratory (Mueller manoeu‐ vre) effort at or near the residual volume, maintained for at least 1–1.5 seconds. These tests are volitional and require patient's full cooperation. For these reasons, this technique is usually performed by an experienced health professional to assure adequate instruction and encour‐ agement. The manoeuver is repeated until the variation between measures is less than 20%. Pimax is usually expressed in absolute values (cmH2O) or as a percentage of the predicted

Pimax is a simple, low-cost and well tolerated technique. In addition, through the hand-held and portable electronic pressure transducers, the technique is easily used by physiotherapists in a wide range of clinical settings (e.g., primary care, hospital wards, intensive care units). Another advantage of using this technique is the availability of reference values for healthy

The reliability of this measure is well established in people with non-CF bronchiectasis [129] and healthy people [130]. In patients with CF, only one study was found assessing the test– retest reliability of Pimax [131]. Pimax had an excellent coefficient of reliability (89%) and

children, adults and the elderly, enabling the interpretation of results [127, 128].

=0.62) exist [120] and support the usability of this

skinfold thickness and 4-C models (R2

52 Cystic Fibrosis in the Light of New Research

*2.2.4. Inspiratory muscle strength*

values.

Inspiratory muscles, in addition to developing strength, have to be able to sustain muscular tasks over time – also known as endurance [126]. Inspiratory muscle endurance is a highly complex ability, which provides insight about the resistance to fatigue of inspiratory muscles and about the function of the inspiratory pump.

Dyspnoea, one of the primary complaints of patients with CF, has been related to inspiratory muscle fatigue. In adults with CF (n=18), inspiratory muscle endurance was found to be strongly correlated with exercise dyspnoea (r=−0.72) and explained 48% of the variability of this symptom [134]. In patients with advanced CF, the assessment and monitoring of inspir‐ atory muscle endurance may have a greater importance since, in these patients, activities of daily living may be limited by their ability to sustain ventilation. Therefore, the measure of inspiratory muscle endurance seems to be useful to evaluate the determinants of dyspnoea and fatigue in patients with CF.

A number of distinct techniques have been employed to measure endurance of the inspiratory muscles [126]. In CF, two main techniques have been employed: (i) ventilatory endurance tasks [43, 112, 131] and (ii) endurance to external loads [132, 134].

In ventilatory endurance tasks, inspiratory muscle endurance has been measured through the sustained maximum inspiratory pressure (SMIP). This sustained pressure is determined using an electronic manometer, with a fixed leak via a 2 mm diameter during the inspiratory manoeuver, and a specific computer software. The leak sets a maximum flow during the inspiratory effort and allows continuous measurement of pressure over a full range of lung volumes, until no further pressure can be generated. During this technique, patients are asked to take a maximal and sustained inspiratory effort from residual volume to total lung capacity. SMIP is measured as the area under the pressure–time curve and is generally expressed in absolute values (pressure–time units) [43, 112, 131].

The coefficient of reliability for measurements of SMIP in CF was previously established as 90% [131]. Albinni et al. (2004) reported that inspiratory muscle endurance, measured by the SMIP, improved significantly in patients with CF (n=16) after 12 weeks of inspiratory muscle training (p=0.0002) [135]. Enright et al. (2004) investigated the effect of an 8-week inspiratory muscle training with a training intensity of either 80% of maximal inspiratory effort (n=9) or 20% of maximal inspiratory effort (n=10) [43]. The SMIP improved significantly in the two groups (from 654–782 to 808–923 pressure–time units; Cohen's dz 0.46 and 0.51) [43].

In endurance to external loads, inspiratory muscle endurance has been determined with incremental loading tests using threshold devices. During these tests, patients have to generate sufficient inspiratory pressure to open the valve and allow inspiratory flow. The test starts with an inspiratory load of 20–30% of Pimax for 2 min. The load is then increased every 2 min in increments of 10% of Pimax. The maximal load is defined by the maximal inspiratory pressure sustained for 1 or 2 min (Plim), which can be expressed in absolute values and as a percentage of the Pimax [132, 134].

The reliability of the Plim in CF has not yet been explored. In patients with COPD and in healthy people, however, it has been demonstrated that the reproducibility of the inspiratory pressure of the threshold was excellent, with small coefficients of variation in both groups (<1%) [136]. Future studies should assess the reliability and test–retest reliability of the Plim in the CF population.

Only one small study was found using Plim as an outcome measure in CF population. In this study, patients with CF (n=7) were submitted to a 6-week inspiratory muscle training. Plim, expressed as a percentage of the Pimax, increased significantly from 49% to 66% (Cohen's dz 1.29) [132]. Using Plim, inspiratory muscle training was also found to have large effect sizes in patients with COPD (n=16; Cohen's dz 0.83) [137] and with chronic heart failure (n=16; Cohen's dz 1.09) [138]. Further research is needed to assess the sensitivity and responsiveness of Plim to inspiratory muscle training in patients with CF.

Both SMIP and Plim appear to be adequate outcome measures to assess the effectiveness of inspiratory muscle training in the inspiratory endurance of patients with CF. Nevertheless, this evidence emerges from few studies with small sample sizes. More research is needed to determine the responsiveness of these two outcome measures to inspiratory muscle training, as well as to other respiratory interventions, such as respiratory retraining.
