Physical training for cystic fibrosis (Review)

Transcription

1 Bradley JM, Moran F This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2008, Issue 1

2 T A B L E O F C O N T E N T S HEADER ABSTRACT PLAIN LANGUAGE SUMMARY BACKGROUND OBJECTIVES METHODS RESULTS DISCUSSION Figure Figure AUTHORS CONCLUSIONS REFERENCES CHARACTERISTICS OF STUDIES DATA AND ANALYSES Analysis 1.1. Comparison 1 Aerobic training versus no physical training, Outcome 1 Change in VO2 max during maximal exercise test (ml/kg/min) Analysis 1.2. Comparison 1 Aerobic training versus no physical training, Outcome 2 Annual rate change in VO2max during maximal exercise test (ml/kg/min) Analysis 1.3. Comparison 1 Aerobic training versus no physical training, Outcome 3 Annual rate decline peak heart rate (beats/min) Analysis 1.4. Comparison 1 Aerobic training versus no physical training, Outcome 4 Change in saturation during maximal exercise test (%) Analysis 1.5. Comparison 1 Aerobic training versus no physical training, Outcome 5 Annual rate of change in VE (L/min) Analysis 1.6. Comparison 1 Aerobic training versus no physical training, Outcome 6 Annual rate change in peak working capacity during maximal exercise test (%) Analysis 1.7. Comparison 1 Aerobic training versus no physical training, Outcome 7 Change in FEV1(%) Analysis 1.8. Comparison 1 Aerobic training versus no physical training, Outcome 8 Change in strength (Newton metres) Analysis 1.9. Comparison 1 Aerobic training versus no physical training, Outcome 9 Annual rate of change in FEV1 (%). 33 Analysis Comparison 1 Aerobic training versus no physical training, Outcome 10 Change FVC (%) Analysis Comparison 1 Aerobic training versus no physical training, Outcome 11 Annual rate of change in FVC (%) Analysis Comparison 1 Aerobic training versus no physical training, Outcome 12 Annual rate of change in FEF (%) Analysis Comparison 1 Aerobic training versus no physical training, Outcome 13 Change in quality of life Analysis Comparison 1 Aerobic training versus no physical training, Outcome 14 Change in weight (kg) Analysis Comparison 1 Aerobic training versus no physical training, Outcome 15 Change in fat free mass (kg). 36 Analysis Comparison 1 Aerobic training versus no physical training, Outcome 16 Annual change of ideal weight for height (%) Analysis Comparison 1 Aerobic training versus no physical training, Outcome 17 Change in activity Analysis 2.1. Comparison 2 Anaerobic training versus no physical training, Outcome 1 Change in VO2 max (ml/kg/min). 38 Analysis 2.2. Comparison 2 Anaerobic training versus no physical training, Outcome 2 Change in lactate during maximal test (mmol/l) Analysis 2.3. Comparison 2 Anaerobic training versus no physical training, Outcome 3 Change in saturation during maximal exercise test (%) Analysis 2.4. Comparison 2 Anaerobic training versus no physical training, Outcome 4 Change in peak power during maximal test (watts) Analysis 2.5. Comparison 2 Anaerobic training versus no physical training, Outcome 5 Change in mean power during maximal test (watts) i

3 Analysis 2.6. Comparison 2 Anaerobic training versus no physical training, Outcome 6 Change in maximum workload during maximal test (watts) Analysis 2.7. Comparison 2 Anaerobic training versus no physical training, Outcome 7 Change in lower limb strength (Newton metres) Analysis 2.8. Comparison 2 Anaerobic training versus no physical training, Outcome 8 Change in FEV1 (%) Analysis 2.9. Comparison 2 Anaerobic training versus no physical training, Outcome 9 Change in FVC (%) Analysis Comparison 2 Anaerobic training versus no physical training, Outcome 10 Change in physical function (CF questionnaire) Analysis Comparison 2 Anaerobic training versus no physical training, Outcome 11 Change in quality of life.. 43 Analysis Comparison 2 Anaerobic training versus no physical training, Outcome 12 Change in weight (kg).. 43 Analysis Comparison 2 Anaerobic training versus no physical training, Outcome 13 Change in fat free mass (kg). 44 Analysis Comparison 2 Anaerobic training versus no physical training, Outcome 14 Change in activity Analysis 3.1. Comparison 3 Combined aerobic and anaerobic training, Outcome 1 Annual change in peak heart rate during constant load bicycle ergometry (beat/min) Analysis 3.2. Comparison 3 Combined aerobic and anaerobic training, Outcome 2 Annual change in peak heart rate during constant load arm ergometry (beat/min) Analysis 3.3. Comparison 3 Combined aerobic and anaerobic training, Outcome 3 Annual change in Ve (L/min) during constant load bicycle ergometry Analysis 3.4. Comparison 3 Combined aerobic and anaerobic training, Outcome 4 Annual change in Ve (L/min) during constant load arm ergometry Analysis 3.5. Comparison 3 Combined aerobic and anaerobic training, Outcome 5 Annual change in lactate (mmol/l)during constant load bicycle ergometry Analysis 3.6. Comparison 3 Combined aerobic and anaerobic training, Outcome 6 Annual change in lactate (mmol/l) during constant load arm ergometry (beat/min) Analysis 3.7. Comparison 3 Combined aerobic and anaerobic training, Outcome 7 Annual change in RER during constant load bicycle ergometry Analysis 3.8. Comparison 3 Combined aerobic and anaerobic training, Outcome 8 Annual change in RER during constant load arm ergometry Analysis 3.9. Comparison 3 Combined aerobic and anaerobic training, Outcome 9 Annual change in RR during constant load bicycle ergometry (breaths/min) Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 10 Annual change in RR during constant load arm ergometry (breaths/min) Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 11 Annual change in Borg breathlessness during constant load bicycle ergometry Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 12 Annual change in Borg breathlessness during constant load arm ergometry Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 13 Annual change in Borg muscle effort during constant load bicycle ergometry Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 14 Annual change in Borg muscle effort during constant load arm ergometry Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 15 Annual change in FEV1 (ml). 52 Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 16 Annual change in FVC(ml).. 52 Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 17 Annual change in BMI (kg/ml). 53 WHAT S NEW HISTORY CONTRIBUTIONS OF AUTHORS DECLARATIONS OF INTEREST DIFFERENCES BETWEEN PROTOCOL AND REVIEW INDEX TERMS ii

4 [Intervention Review] Judy M Bradley 1, Fidelma Moran 1 1 Health and Rehabilitation Sciences Research Institute and School of Health Sciences, University of Ulster, Newtownabbey, UK Contact address: Judy M Bradley, Health and Rehabilitation Sciences Research Institute and School of Health Sciences, University of Ulster, Shore Road, Newtownabbey, Northern Ireland, BT37 0QB, UK. [email protected]. Editorial group: Cochrane Cystic Fibrosis and Genetic Disorders Group. Publication status and date: Edited (no change to conclusions), published in Issue 7, Review content assessed as up-to-date: 7 March Citation: Bradley JM, Moran F.. Cochrane Database of Systematic Reviews 2008, Issue 1. Art. No.: CD DOI: / CD pub2. Background A B S T R A C T Physical training may form an important part of the care package for people with cystic fibrosis. Objectives To determine whether a prescribed regimen of physical training produces improvement or prevents deterioration in physiological and clinical outcomes in cystic fibrosis compared to no training. Search methods We searched the Cochrane Cystic Fibrosis and Genetic Disorders Group Trials Register which comprises references identified from comprehensive electronic database searches and handsearches of relevant journals and abstract books of conference proceedings. Date of the most recent search: 29 November Selection criteria All randomised and quasi-randomised controlled clinical trials in which a prescribed regimen of physical training is compared to no physical training in people with cystic fibrosis. Data collection and analysis Two authors independently selected studies for inclusion, assessed methodological quality and extracted data. Main results Of the 28 studies identified, seven studies which included 231 participants, met the inclusion criteria. This review does provide some limited evidence from both short- and long-term studies that aerobic or anaerobic physical training has a positive effect on primary outcomes (exercise capacity, strength and lung function) but improvements are not consistent between studies. Authors conclusions Conclusions about the efficacy of physical training in cystic fibrosis are limited by the small size, short duration and incomplete reporting of most of the studies included in this review. Physical training is already part of the care package offered to most people with cystic fibrosis and there is a lack of evidence to actively discourage this. The benefits obtained from including physical training in a package of care may be influenced by the type of training programme. Further research is needed to assess comprehensively the benefits of exercise 1

5 programmes in people with cystic fibrosis and the relative benefits of the addition of aerobic versus anaerobic versus a combination of both types of physical training to the care of people with cystic fibrosis. P L A I N L A N G U A G E S U M M A R Y Physical training to improve exercise capacity in people with cystic fibrosis The progress of lung disease in cystic fibrosis leads to abnormal breathing during exercise. This limits people exercising, which in turn affects health and body image. Physical training is designed to improve physical, heart and muscle strength through aerobic and anaerobic activity. Aerobic exercise involves training for a length of time such as distance cycling or running. Anaerobic exercise involves training intensely for a short time such as weight training or sprinting. A lack of regular physical training may lead to more severe lung disease and a reduced ability to perform day-to-day tasks. The side effects of exercise include dehydration, damage to muscles and bone fractures in those with low bone mineral density. This review includes seven studies with 231 participants. Due to different study designs, we were not able to combine any data. Some studies were short-term and did not show differences between treatments. However, it is not likely that training for less than one month would be beneficial. There is some evidence that physical training produces benefits and that these are influenced by the type of training programme. There are not enough studies in this review to compare aerobic training to anaerobic training or to a mixture of these two. The review provides some evidence that the benefits of training are maintained after training has ceased, but it is not clear for how long. Physical training is already part of the care package offered to most people with CF and there is no evidence to discourage this. B A C K G R O U N D Description of the condition Cystic fibrosis (CF) is the most common life-limiting autosomal recessively inherited disease in Caucasian populations, with a carrier rate of 1 in 25 and an incidence of 1 in 3500 live births in the USA (CF Foundation 2009, CF Trust 2010). Although this is a multisystem disease, the primary cause of death is respiratory failure (CF Trust 2010). Progressive respiratory disease results in an abnormal ventilatory response to exercise in CF. This contributes to dyspnoea (shortness of breath) and is a major limiting factor to exercise tolerance in this population (O Neill 1987). Description of the intervention Physical training is defined as participation in a programme of regular vigorous physical activity designed to improve physical performance or cardiovascular function or muscle strength or any combination of these three (Shephard 1994). There are basically two different types of physical training: aerobic training or anaerobic training. Aerobic training usually involves periods of continuous training for a length of time at a target intensity (eg cycling or running). Anaerobic training involves training at a high intensity for a short duration (eg weight or resistance training or sprinting). How the intervention might work It is accepted by the majority of people with CF and healthcare professionals that physical training is important in CF. A recent survey of specialist UK CF centres highlighted the availability of physical training facilities varies widely across CF centres (Lloyd 2010). Adherence to physical training contributes to the alleviation of dyspnoea and reduced exercise tolerance in CF. Physical training may also be an important part of the management of diabetes in CF, as exercise improves appetite and contributes to a more positive body image (Peebles 1998). Physical training maintains pulmonary function by improving sputum clearance and reducing residual volume (O Neill 1987). Physical training may delay the onset of osteoporosis by preventing a reduction in bone mineral density (Wolman 1995). Other postulated benefits of any physical training may be decreased anxiety and depression, enhanced feelings of well-being and enhanced performance at work, recreational and sport activities (ACSM 2010). Finally, adherence to physical training is important because exercise capacity may be an independent prognostic risk factor in CF (Nixon 1992). It is not clear how many weeks training are required to achieve these benefits or what combination of aerobic and anaerobic training is required but there is usually a drop of one per cent exercise capacity per day seven days after stopping exercise (McArdle 2001). Nonadherence to prescribed physical training may contribute to worsening signs and symptoms of respiratory disease and a reduced 2

6 ability to perform activities of daily living and thus ultimately have a detrimental affect on the individual s prognosis. The side effects of physical training include dehydration, musculoskeletal damage and fractures in those with low bone mineral density (Mazzeo 1998). Why it is important to do this review It is therefore essential to clearly establish the net effect of physical training in order to provide a strong evidence base to support the inclusion of physical training in the demanding care package for people with CF and to maximise adherence to prescribed physical training programmes. O B J E C T I V E S To determine whether a prescribed regimen of physical training produces improvement or prevents deterioration in physiological and clinical outcomes in CF compared to no training. M E T H O D S Criteria for considering studies for this review Types of studies Randomised controlled trials (RCT) or quasi-randomised controlled clinical trials. Types of participants People with CF, of any age, and any degree of disease severity, diagnosed on the basis of clinical criteria and sweat testing or genotype analysis. Specific details on age and degree of disease severity at commencement of the study were recorded. Types of outcome measures Primary outcomes 1. Exercise capacity i) objective change in exercise capacity (all exercise tests) ii) objective change in other physiological indices of exercise capacity 2. Measures of specific indices of strength, mass, effort and general fatigue 3. Pulmonary function tests - the change in per cent predicted or/and absolute change from baseline compared to control for the following measures (If other parameters have been used, they have been considered): i) forced expiratory volume in one second (FEV 1 ); ii) forced vital capacity (FVC); iii) forced expiratory flows between 25-75% of expired volume (FEF ); iv) total lung capacity (TLC); v) functional residual capacity (FRC); The fourth primary outcome mortality was moved to Secondary outcomes in line with Cochrane Collaboration guidance to limit the number of primary outcomes to three. Secondary outcomes 1. Mortality 2. Symptom scores - any objective changes in symptom scores 3. Quality of life - all quality of life instruments used 4. Weight - all validated measures of weight and nutrition 5. Number of acute exacerbations, intravenous antibiotic courses and time off work or school 6. Measures of oxygenation, either by oximetry or blood gas analysis 7. Measures of bone mineral density and diabetic control 8. Compliance with physical training 9. Compliance with other treatment, such as chest physiotherapy, nutritional regimens 10. Adverse effects including fractures, skeletal muscle injuries and death 11. Cost evaluation Types of interventions Any type of prescribed physical training. The precise nature of the training (intensity, frequency and duration) has been recorded. Any studies, which do not include a prescribed physical exercise component, have been excluded. As the aim of this review was to assess the efficacy of physical training, studies which involve respiratory muscle training exclusively, have been excluded. Important alterations in patient management such as other physical therapies (eg airway clearance techniques) or medication must be controlled for. Search methods for identification of studies Electronic searches Relevant studies were identified from the Group s Cystic Fibrosis Trials Register using the term: exercise. The Cystic Fibrosis Trials Register is compiled from electronic searches of the Cochrane Central Register of Controlled Trials 3

7 (Clinical Trials) (updated each new issue of The Cochrane Library), quarterly searches of MEDLINE, a search of EMBASE to 1995 and the prospective handsearching of two journals - Pediatric Pulmonology and the Journal of Cystic Fibrosis. Unpublished work is identified by searching through the abstract books of three major cystic fibrosis conferences: the International Cystic Fibrosis Conference; the European Cystic Fibrosis Conference and the North American Cystic Fibrosis Conference. For full details of all searching activities for the register, please see the relevant sections of the Cystic Fibrosis and Genetic Disorders Group Module. Date of the most recent search of the Group s Cystic Fibrosis Trials Register: 29 November Searching other resources The reference lists of each RCT were searched for additional publications that may contain RCTs. Authors of studies in this review were contacted for further study details or for information on other published and unpublished studies. Three authors responded: one author confirmed that data extracted were correct and that no further data were available (Cerny 1989); a further author stated that study data were not available (Michel 1989); and a third author stated that they were in the process of writing up the abstract for publication but that this is not a randomized controlled study (Hebestreit 2003). Data collection and analysis degree of blinding in the study, adverse effects, inclusion and exclusion criteria, dropouts or withdrawals, whether intention-totreat analyses were undertaken and whether the method of statistical analysis described from the study was recorded. Authors also examined for selective reporting and any other potential sources of bias. Methods of randomisation and allocation concealment were judged to have either a low, unclear or high risk of bias depending on the methods used. With regards to blinding, the potential risk of bias was deemed to increase as the number of people blinded to an intervention decreased. The authors also considered unexplained dropouts from a study or an unequal number of dropouts across treatment groups to pose a potential risk of bias. A lack of other information, e.g. on adverse effects, statistical methods etc, could also constitute a potential risk of bias. Measures of treatment effect For continuous outcomes, where between-group differences in the mean change from baseline were recorded, the authors calculated the mean difference (MD) if possible. When data on the standard deviation (SD) for each group were not available, but data for the standard error (SE) of the difference were, the authors sought to use the generic inverse variance (GIV). Where possible, they have taken the published standard error of the mean (SEM), but where this was not available, they have used the published confidence intervals to estimate a SE. If dichotomous outcomes had been identified they would have been analysed using the relative risk or odds ratio. Selection of studies Two authors (JB, FM) independently selected the studies to be included in the review. Excluded studies included non-rcts, those studies involving respiratory muscle training exclusively and those which did not have a programme of physical training. If disagreement arose on the suitability of a study for inclusion in the review the authors reached a consensus by discussion. The authors recorded any areas of disagreement. Data extraction and management Each author independently extracted data using standard data acquisition forms. If disagreement arose on the quality of a study the authors reached a consensus by discussion. The authors recorded any areas of disagreement. Assessment of risk of bias in included studies In order to establish a risk of bias for each study, two authors (JB, FM) independently assessed the studies according to the Cochrane risk of bias tool (Higgins 2009). In particular authors examined details of the randomisation method, allocation concealment, the Unit of analysis issues We have not included any cross-over trials in this latest version of the review. Dealing with missing data Authors of studies in this review were contacted for further study details where necessary. Assessment of heterogeneity The authors were unable to combine data for any of their listed outcomes; however, if for future updates of this review they are able to combine any data, they plan to test for heterogeneity between studies using the I 2 method (Higgins 2003). The authors will assess heterogeneity as low if the value for I 2 is around or below 25%, as moderate if it is around 50% and high if it is over 75%. Assessment of reporting biases We assessed selective reporting (comparing methods sections and results sections from the included papers) and other relevant bias. These were summarised into risk of bias tables and figures. 4

8 Data synthesis The authors will use a random-effects model or a fixed effect model depending on the level of heterogeneity observed. Subgroup analysis and investigation of heterogeneity The authors plan to investigate the causes of heterogeneity by subgroup analysis (children versus adults). Sensitivity analysis In order to investigate whether heterogeneity impacted upon the overall pooled effect estimate, the authors plan to apply randomeffects modelling, and compare this with a fixed effect model. They also plan a sensitivity analysis with and without quasi-randomised studies. R E S U L T S Description of studies See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of ongoing studies. Results of the search Of the 28 studies identified, seven studies met the inclusion criteria, 20 studies were excluded and one study is ongoing. Included studies Of the 28 studies identified seven studies, which included 231 participants, met the inclusion criteria (Cerny 1989; Klijn 2004; Michel 1989; Moorcroft 2004; Schneiderman 2000; Selvadurai 2002; Turchetta 1991). All included studies were of randomised parallel group design. All seven studies included a control group which did not receive a prescribed exercise programme. As the aim of this review was to assess the efficacy of physical training, studies which involved respiratory muscle training exclusively were excluded. One study compared two types of physical training programmes (aerobic training or anaerobic training) with a control group (Selvadurai 2002). Another study compared anaerobic training to a control group (Klijn 2004). Four studies compared only aerobic training to a control group (Cerny 1989; Michel 1989; Schneiderman 2000; Turchetta 1991). One study compared the effects of a combined training programme to a control group (Moorcroft 2004). Five of the seven studies were published in full (Cerny 1989; Klijn 2004; Moorcroft 2004; Schneiderman 2000; Selvadurai 2002). One study included adults only (Moorcroft 2004). Three studies included children only (Klijn 2004; Selvadurai 2002; Turchetta 1991) and three studies included both adults and children (Cerny 1989; Michel 1989; Schneiderman 2000). Studies included participants with a range of disease severity. Four studies were of short duration (less than one month) and were carried out during hospital admissions (Cerny 1989; Michel 1989; Selvadurai 2002; Turchetta 1991). In one study the hospital admission was for routine assessment (Turchetta 1991); in two other studies the hospital admission was for an acute exacerbation (Cerny 1989; Selvadurai 2002); and in a further study the reason for admission was not detailed and the duration was not reported (Michel 1989). The other three longer-term studies (more than one month) were outpatient-based (Klijn 2004; Moorcroft 2004; Schneiderman 2000). Klijn was a three-month study with a threemonth follow up (Klijn 2004); Moorcroft was a one-year study (Moorcroft 2004); and the Schneiderman study lasted three years (Schneiderman 2000). With the exception of the Cerny study, all groups were similar at baseline (Cerny 1989). Cerny reported a significant difference in FEV 1 and FEF between the groups (Cerny 1989). Participants were followed up in three studies and reasons for loss to follow up were reported (Klijn 2004; Michel 1989; Selvadurai 2002). In one study the number of participants in each group was not detailed, the mean difference between the treatment and control group could not be calculated (Michel 1989). Excluded studies Twenty studies were excluded for the reasons which follow: 11 studies were not RCTs (Andreasson 1987; Barry 2001; de Jong 1994; Edlund 1986; Hebestreit 2003; Heijerman 1992; Kriemler 2001; Orenstein 1981; Salh 1989; Stanghelle 1998; Tuzin 1998); seven studies did not include a physical training programme as per our protocol (Albinni 2004; Aquino 2006; Balestri 2004; Bilton 1992; Chatham 1997; Falk 1988; Lannefors 1992); and two studies did not use a control arm with no physical training (Hall 2010; Oreinstein 2004). Risk of bias in included studies We attempted to establish a risk of bias for the following domains by assessing the methodological quality of each study based on the Cochrane risk of bias tool (Higgins 2009). Allocation Three studies described the methods used for generation of the randomisation sequence and these were judged to be adequate leading to a low risk of bias for these studies (Moorcroft 2004; 5

9 Schneiderman 2000; Selvadurai 2002). Four studies were described as randomised but did not give any details of the methods used, these were deemed to have an unclear risk of bias (Cerny 1989; Klijn 2004; Michel 1989; Turchetta 1991). Only two studies described how allocation was adequately concealed and these were judged to have a low risk of bias (Klijn 2004; Selvadurai 2002). Five studies did not give any details of the method of allocation concealment and were deemed to have an unclear risk of bias (Cerny 1989; Michel 1989; Moorcroft 2004; Schneiderman 2000; Turchetta 1991). Blinding None of the studies were blinded and hence were subject to a potential risk of bias; however, the studies should be considered in the context of the difficulty of blinding these types of studies. It would not have been possible to blind the participants to physical training interventions. It is unclear whether all the outcome measures were performed by assessors not involved in the delivery of the interventions or by the people involved in the delivery of the interventions. Incomplete outcome data Five studies gave a clear description and details about dropouts (Klijn 2004; Moorcroft 2004; Schneiderman 2000; Selvadurai 2002; Cerny 1989). Moorcroft also stated the number of participants excluded for missing more than one assessment session (Moorcroft 2004). These studies were judged to have a low risk of bias. Two studies did not give any details of dropouts or whether intention-to-treat analysis had been used and were judged to have an unclear risk of bias (Michel 1989; Turchetta 1991). (Klijn 2004; Moorcroft 2004; Selvadurai 2002). Two further studies stated the inclusion criteria but not the exclusion criteria which could be a potential source of bias (Cerny 1989; Schneiderman 2000). Five studies clearly described the methods of statistical analysis, thus eliminating a potential source of bias (Cerny 1989; Klijn 2004; Moorcroft 2004; Schneiderman 2000; Selvadurai 2002). Effects of interventions Where primary studies reported statistical or non-statistical differences between groups but did not provide adequate data (means and standard deviations (SD)) that could be presented in the RevMan software (Review Manager 2008), the information from the primary (original) study has been included in the results. It was not possible to pool data for any outcomes due to variations in the type and duration of studies, the times at which outcomes were measured, the different methods of reporting outcomes, the omission of data relating to either mean change from baseline for each group and the SD or SE. Where the SD for each group was not available, but the SE of the difference was available, the authors have analysed the data using the GIV method. The results have been reported under effects of aerobic training, anaerobic training and combined aerobic and anaerobic training and further subdivided into the short and longer-term studies. There are four short-term aerobic studies (Cerny 1989; Michel 1989; Selvadurai 2002; Turchetta 1991); one long-term aerobic study (Schneiderman 2000); one short-term anaerobic study ( Selvadurai 2002); one long-term anaerobic study (Klijn 2004); and one long-term combined aerobic and anaerobic training study (Moorcroft 2004). Selective reporting In four studies all outcomes detailed in methods were reported in results and data reported for all time-points. (Cerny 1989; Klijn 2004; Schneiderman 2000; Moorcroft 2004). These studies were judged to have a low risk of bias. Two studies were in abstract format and so it was unclear if outcomes detailed in methods were reported in results and data were reported for all time-points. They were deemed to have an unclear risk of bias (Michel 1989; Turchetta 1991). In one study data were not reported for all outcomes detailed in the methods and this was deemed to have a high risk of bias (Selvadurai 2002). Other potential sources of bias Two of these studies are only available in abstract format and do not state inclusion or exclusion criteria, nor do they describe the methods of statistical analysis used which could be a source of bias (Michel 1989; Turchetta 1991). Three studies clearly stated inclusion and exclusion criteria which limits the potential for bias Aerobic versus no physical training Primary outcomes 1. Exercise capacity a. Objective change in exercise capacity This outcome was reported in two studies (Schneiderman 2000; Selvadurai 2002). In the study by Selvadurai, exercise capacity was measured by peak oxygen uptake (VO 2 peak) during a treadmill exercise test (Selvadurai 2002). In this study, improvements in exercise tolerance during aerobic training were significantly greater than with no specific training, MD 8.53 ml/kg/min (95% confidence interval (CI) 4.85 to 12.21). 6

10 In the study by Schneiderman, exercise capacity was measured during cycle ergometry (Schneiderman 2000). In this study there was no significant difference in the annual rate of decline in VO 2 peak between groups, MD 0.05 ml/kg/min (95% CI to 1.20) (Schneiderman 2000). b. Other physiological indices of exercise capacity These outcomes were reported in three studies (Cerny 1989; Schneiderman 2000; Selvadurai 2002). In the study by Cerny exercise tolerance was measured during cycle ergometry (Cerny 1989). Two studies reported on heart rate (Cerny 1989; Schneiderman 2000). Cerny reported that there was no significant difference between control and treatment arms in change in peak heart rate or the ratio of peak heart rate to peak load (Cerny 1989). In the Schneiderman study, it was reported that there was no significant difference in the annual rate of decline in peak heart rate between groups, MD 1.10 beats per minute (bpm) (95% CI to 3.04) (Schneiderman 2000). One study reported on desaturation during exercise (Selvadurai 2002). The aerobic training group demonstrated less desaturation following training compared to control, MD 0.62% (95% CI 0.32 to 0.92). These differences did not reach statistical significance in the original study (Selvadurai 2002). Two studies reported on minute ventilation (Cerny 1989; Schneiderman 2000). The study by Cerny states that there were no differences between groups in change in ratio peak minute ventilation/peak load (Cerny 1989). In the study by Schneiderman there was no significant difference in the annual rate of decline in maximum minute ventilation between groups, MD 2.09 L/min (95% CI to 5.77) (Schneiderman 2000). One study reported on peak working capacity (Schneiderman 2000). This study states that there was no significant difference in the annual rate of decline in peak working capacity between groups, MD 0.82% (95% CI to 3.55) (Schneiderman 2000). 2. Specific Indices of strength, mass effort and general fatigue This outcome was reported in only one study (Selvadurai 2002). In this study the aerobic training group had a significantly greater increase in lower limb strength than the control group, MD 8.13 Newton metres (Nm) (95% CI 4.49 to 11.77) (Selvadurai 2002). 3. Pulmonary function tests This outcome was reported in three studies (Cerny 1989; Schneiderman 2000; Selvadurai 2002). Two aerobic training studies reported on the short-term changes in lung function (less than one month) (Cerny 1989; Selvadurai 2002). In the study by Cerny there was no difference in the change in FEV 1 percent predicted or FVC per cent predicted. In the study by Selvadurai there was no significant difference between the groups in increase in FEV 1 per cent predicted, MD 2.03% (95% CI to 6.37) or FVC per cent predicted, MD 0.06% (95% CI to 2.67). Schneiderman reported on the effects of aerobic physical training on lung function at three years (Schneiderman 2000). The control group demonstrated a significantly greater mean rate of annual decline in FVC per cent predicted than the exercise group, MD 2.17% (95% CI 0.47 to 3.87). There was a similar trend reported for the outcomes FEV 1 per cent predicted, MD 2.01% (95% CI to 4.08) and FEF per cent predicted, MD 0.80% (95% CI to 3.80), although this was not statistically significant (Schneiderman 2000). Secondary outcomes 1. Mortality No data were reported from any of the studies. 2. Symptom scores This outcome was reported in one of the studies (Cerny 1989). The study reported no differences between the groups in the number of coughs or in dry sputum weight or volume, but no data were provided (Cerny 1989). 3. Quality of life This outcome was reported in two studies (Schneiderman 2000; Selvadurai 2002). Schneiderman reported on attitudes toward physical activity and perceived feasibility of a regular aerobic exercise programme (Schneiderman 2000). Positive effects, reported by 43 out of 49 participants, included feeling better about themselves, having more energy and less chest congestion. A small number of participants reported no differences. Both groups stated it would be feasible to meet aerobic exercise targets, if requested to do so by doctors (Schneiderman 2000).Selvadurai assessed quality of life using the Quality of Well-being Scale. Since this scale was previously only validated in an outpatient setting, assessment was undertaken on the participants admission to hospital and one month after their discharge. There was a significant difference between the groups in the change in quality of life MD 0.10 (95% CI 0.03 to 0.17) (Selvadurai 2002). 4. Weight This outcome was reported in two studies (Schneiderman 2000; Selvadurai 2002). 7

11 In the study by Selvadurai there was no difference in change in weight, MD kg (95% CI to 0.13); or change in fat-free mass, MD 0.01 kg (95% CI to 0.21) in the aerobic training group compared with the control group (Selvadurai 2002). In the study by Schneiderman there was no significant difference in the annual rate of decline in per cent of ideal weight for height, MD 0.52% (95% CI to 1.80) (Schneiderman 2000). 5. Number of acute exacerbations in hospital One study reported on this outcome (Schneiderman 2000). There was no significant difference between groups for mean number of hospitalisations or mean number of days in hospital at year one, two and three (Schneiderman 2000). 6. Oxygen saturation Only the Selvadurai study reported on oxygen desaturation during exercise testing (Selvadurai 2002) (See outcome (1) (b) Other physiological indices of exercise capacity ). 7. Measures of bone mineral density and diabetes No data were reported in any of the studies. 8. & 9. Compliance with physical training and other treatment One study reported on this outcome (Schneiderman 2000). In this study, mean scores for compliance with exercise (possible range zero to two indicating poor, partial or full compliance respectively) within the exercise group for year one (1.51), year two (1.51) and year three (1.49) were not significantly different. These scores were always higher than the scores for compliance with airway clearance techniques. Compliance with airway clearance was not statistically different between the groups (Schneiderman 2000). Follow up after physical training concluded Two studies reported follow-up results after termination of formal physical training at one month post-training (Michel 1989; Selvadurai 2002). Michel studied the effects on lung function, but no data were reported (Michel 1989). The increase in weight in the aerobic exercise group (6.4 +/- 4.8 lbs) was greater than in the non-exercise group (3.8 +/- 3.4 lbs) at one-month follow up. Michel also reported that there was a trend towards a greater increase in the sum of four skin folds in the exercise group and mid-arm muscle circumference than the non-exercise group (Michel 1989). These data cannot be entered into the data tables, as the number of participants assigned to each treatment group was not reported. We have contacted the authors for further information, but there is none available. In the Selvadurai study, any increase in study parameters as identified in the above results was maintained at one month. In addition, Selvadurai reported on changes in activity levels by using an accelerometer and an activity diary (Selvadurai 2002). There were no differences in the levels between the aerobic training group compared to control MD 1.20 (95% CI to 2.60) though in the original study this difference was statistically significant (Selvadurai 2002). 10. Adverse effects including fractures, skeletal muscle injuries and death No data were reported in any of the studies. 11. Cost evaluation No data were reported in any of the studies. Anaerobic training versus no physical training Primary outcomes 1. Exercise capacity a. Objective change in exercise capacity This outcome was reported in two studies (Klijn 2004; Selvadurai 2002). Results could not be combined in a meta-analysis since the outcome was assessed at different time-points.in the study by Selvadurai anaerobic training was not associated with improvements in VO 2 max compared with control at discharge, MD 1.95 (95% CI to 5.51) (Selvadurai 2002). In the study by Klijn, exercise capacity was measured during cycle ergometry and the change in VO 2 peak was significantly greater in the anaerobic training versus control group, MD 2.10 ml/kg/ min (95% CI 0.12 to 4.08) (Klijn 2004). b. Other physiological indices of exercise capacity These outcomes were reported in two studies (Klijn 2004; Selvadurai 2002). One study reported on lactate levels (Klijn 2004). The Klijn study reported a significantly greater change in lactate levels in the anaerobic training versus control group, MD 3.40 mmol/l (95% CI 1.33 to 5.47) (Klijn 2004). 8

12 One study reported on desaturation during exercise (Selvadurai 2002). The anaerobic training group demonstrated less desaturation following training compared to control, MD 0.33% (95% CI 0.04 to 0.62). These differences did not reach statistical significance in the original study (Selvadurai 2002). 2. Specific Indices of strength, mass effort and general fatigue These outcomes were reported in two of the studies (Klijn 2004; Selvadurai 2002). The Klijn study reported a significantly greater change in peak power, mean power and maximum work in the anaerobic training versus control group: peak power, MD watts (W) (95% CI to ); mean power, MD W (95% CI to 64.04); and maximum work, MD W (95% CI 4.11 to 21.89) (Klijn 2004). In the study by Selvadurai, the anaerobic training group had a significantly greater increase in lower limb strength than the control group, MD (95% CI to 28.51) (Selvadurai 2002). Selvadurai assessed quality of life using the Quality of Well-being Scale. Since this scale was previously only validated in an outpatient setting, assessment was undertaken on the participants admission to hospital and one month after their discharge. There was no significant difference between the groups in the change in quality of life, MD 0.03 (95% CI to 0.10) (Selvadurai 2002). 4. Weight This outcome was reported in two studies (Klijn 2004; Selvadurai 2002). In the study by Selvadurai there was significantly greater change in weight, MD 1.73 kg (95% CI 1.35 to 2.11) and change in fatfree mass, MD 1.80 kg (95% CI 1.57 to 2.03) in the anaerobic group than the control group (Selvadurai 2002). In the study by Klijn, it was reported within the paper, there was no significant difference in change in body composition between the groups at end of training period (Klijn 2004). 3. Pulmonary function tests This outcome was reported in two studies (Klijn 2004; Selvadurai 2002). In the study by Selvadurai, the anaerobic training group did show a significantly greater mean percentage increase in FEV 1, MD of 5.58% (95% CI 1.34 to 9.82). There was no significant change in percentage increase in FVC in the anaerobic training group, MD 0.17% (95% CI to 2.65), compared to control (Selvadurai 2002). In the Klijn study there were no between group significant differences in lung function (FVC), but no data were reported (Klijn 2004). 5. Number of acute exacerbations in hospital No data were reported from any of the studies. 6. Oxygen saturation Only the Selvadurai study reported on oxygen desaturation during exercise testing (Selvadurai 2002) (See outcome (1) (b) Other physiological indices of exercise capacity ). 7. Measures of bone mineral density and diabetes No data were reported in any of the studies. Secondary outcomes 1. Mortality No data were reported from any of the studies. 2. Symptom scores No data were reported from any of the studies. 3. Quality of life This outcome was reported in two studies (Klijn 2004; Selvadurai 2002). In the Klijn study there was no significant difference in physical function between the groups at the end of the anaerobic training period, MD 1.30 (95% CI to 14.15). No other difference was found in any other quality of life domain (Klijn 2004). 8. & 9. Compliance with physical training and other treatment One study reported on this outcome (Klijn 2004). In this study, the mean attendance rate at exercise sessions was 98.1% (SD 4.3). Reasons for absence were holidays or sickness (Klijn 2004). Follow up after physical training concluded One study reported follow-up results after termination of formal physical training at one month post-training (Selvadurai 2002) and one study at three months post-training (Klijn 2004). In addition, Selvadurai reported on changes in activity levels by using an accelerometer and an activity diary (Selvadurai 2002). There were no differences in the levels between the anaerobic training group compared to control MD 0.65 (95% CI to 1.98) though original study data was significant. The change in quality 9

13 of life score was not significantly different during anaerobic training compared to no training (Selvadurai 2002). In the Klijn study, the domain of physical functioning in the training group was significantly higher compared to pre-training levels at the end of the follow-up period. Also, the increase in the anaerobic training group decreased to baseline levels with the exception of peak power, mean power and quality of life. A subgroup of participants (aerobic n = 5; anaerobic n = 18; control n = 16) who completed an activity diary and wore an activity accelerometer showed no significant differences for between group comparisons in habitual activity at follow up (Klijn 2004). 10. Adverse effects including fractures, skeletal muscle injuries and death No data were reported in any of the studies. 11. Cost evaluation No data were reported in any of the studies. Combined aerobic and anaerobic training versus no training The Moorcroft study showed a significant reduction in lactate during bicycle ergometry mmol/l (95% CI to -0.12), but not during arm ergometry mmol/l (95% CI to 0.50) in the training group compared to the control group (Moorcroft 2004). The Moorcroft study also showed no significant reduction in respiratory exchange ratio (RER) during bicycle ergometry 0.02 (95% CI to 0.06) and no significant reduction during arm ergometry 0.00 (95% CI to 0.04) in the training group compared to the control group (Moorcroft 2004). This study also reported respiratory rate (RR), but showed no significant reduction in RR during bicycle ergometry (95% CI to 3.30) and no significant reduction during arm ergometry, (95% CI -1.9 to 0.10) in the training group compared to the control group (Moorcroft 2004). No significant reduction was shown by the Moorcroft study data for Borg scores: either for breathlessness during bicycle ergometry 0.00 (95% CI to 1.00) and during arm ergometry, (95% CI to 0.10) in the training group compared to the control group; or for muscle fatigue during bicycle ergometry, (95% CI to 0.90) and during arm ergometry, 0.30 (95% CI to 1.59) in the training group compared to the control group (Moorcroft 2004). Primary outcomes 2. Specific Indices of strength, mass effort and general fatigue No data were reported in any of the studies. 1. Exercise capacity a. Objective change in exercise capacity No data were reported in any of the studies. b. Other physiological indices of exercise capacity These outcomes were reported in one study (Moorcroft 2004). Exercise capacity was measured by other physiological indices at the end of an identical constant work rate of 55% of their maximal workload on incremental testing for arm and bicycle ergometry (Moorcroft 2004).The Moorcroft study showed a significant reduction in heart rate in the training group compared to the control group during bicycle ergometry -8.2 bpm (95% CI to ), but not during arm ergometry -1.4 bpm (95% CI to 7.40) (Moorcroft 2004). Moorcroft showed no significant reduction in ventilation during bicycle ergometry -2.5 litres per minute (L/min) (95% CI to 1.11), but a significant reduction during arm ergometry L/min (95%CI to -0.20) in the training group compared to the control group (Moorcroft 2004). 3. Pulmonary function tests This outcome was reported in one study (Moorcroft 2004). In this study, data showed no significant change in FEV 1 (ml) ml (95% CI to ) in the training group compared to the control group; however, these data did show a significant change in FVC (ml) ml (95% CI 3.01 to ) in the training group compared to the control group (Moorcroft 2004). Secondary outcomes 1. Mortality No data were reported in any of the studies. 2. Symptom scores No data were reported in any of the studies. 3. Quality of life No data were reported in any of the studies. 10

14 4. Weight This outcome was reported in one study (Moorcroft 2004). Data showed no significant change in body mass index (BMI) 0.54 kg/m2 (95% CI to 1.17) in the training group compared to the control group (Moorcroft 2004). 5. Number of acute exacerbations in hospital No data were reported in any of the studies. 6. Oxygen saturation No data were reported in any of the studies. 7. Measures of bone mineral density and diabetes No data were reported in any of the studies. 8. & 9. Compliance with physical training and other treatment No data were reported in any of the studies. 10. Adverse effects including fractures, skeletal muscle injuries and death No data were reported in any of the studies. 11. Cost evaluation No data were reported in any of the studies. D I S C U S S I O N This systematic review examined the effectiveness of physical training in the management of CF. Despite a review of available literature; only seven relevant primary studies were included, of which only five were available as full text publications. As outlined in the Results section, pooling of results was not possible. However, this review does provide some limited evidence from both short- and long-term studies that aerobic or anaerobic physical training has a positive effect on primary outcomes (exercise capacity, strength and lung function) but improvements are not consistent between studies. The Cochrane risk of bias tool was used to assess aspects of the methodological quality of the studies in this review (Higgins 2009) and the results are summarised in the Risk of Bias Tables and Figures 1 and 2 (Characteristics of included studies; Figure 1; Figure 2). Overall the risk of bias in these studies is likely to be low for some aspects of quality and unclear for other aspects. All of the studies used random allocation, and three studies provided details on the specific procedures used (Moorcroft 2004; Schneiderman 2000; Selvadurai 2002). Two studies reported their method of allocation concealment (Klijn 2004; Selvadurai 2002). This implies that there was some attempt to ensure groups were comparable by preventing selection bias. None of the studies were double blinded. Single blinding (independent outcome assessment) is a more important quality issue than double blinding in physical training studies. In one study the randomisation used was not successful in ensuring the groups were balanced at baseline in all measures (Cerny 1989). Two studies reported that they had participants lost to follow up (Moorcroft 2004; Schneiderman 2000). 11

15 Figure 1. Methodological quality summary: review authors judgments about each methodological quality item for each included study. 12

16 Figure 2. Methodological quality graph: review authors judgments about each methodological quality item presented as percentages across all included studies. As the studies in this review included recruitment of mixed populations with regard to age, disease severity and stability, this means that the results have some applicability to the general population of people with CF. As regards the physical training intervention, the majority of studies provided information on the type, frequency, intensity and duration of the physical training intervention. According to the American College of Sports Medicine guidelines for physical training programmes for low or deconditioned individuals (ie exercising for a tolerable time and aiming to progress to at least 20 to 30 minutes exercise, at 55% to 64% maximum heart rate, for three to five days a week (ACSM 2010)) the training interventions in all of the studies (with the exception of one, which provides no detail of the intervention (Michel 1989)) was sufficient to achieve a training effect. Although standard outcome measures were used in the studies to assess efficacy of physical training, estimates for the minimal clinically important differences of these outcome measures are not available. While the effect sizes for some of the outcome measures in this review were statistically significant, the clinical significance of these results remains open to interpretation. Four of the seven studies were of short duration (less than one month) (Cerny 1989; Michel 1989; Selvadurai 2002; Turchetta 1991). The length of training required to obtain physiological benefits in CF is not known, but it is unlikely that training for short periods (less than one month) would achieve physiological benefit (Casaburi 1992). The short duration may explain why some of the studies did not detect differences in important primary outcome measures. Two of the short-term studies reported beneficial effects of physical training during an acute exacerbation of respiratory disease (Cerny 1989; Selvadurai 2002). However, there are a number of important factors to consider when prescribing exercise during an acute exacerbation. Physical training may heighten symptoms associated with an acute exacerbation such as bronchospasm, cough, wheeze and dyspnoea. Individuals may show higher oxygen uptake, higher heart rate, higher respiratory rate and increased lung compliance during acute exacerbations (Primos 1996). It is important that exercise is modified and individually tailored according to the individual s symptoms at the time of the exacerbation. It is unclear how much of the improvements demonstrated by physical training during an exacerbation are actual improvements in exercise capacity and how much is reflective of resolution of the acute exacerbation and accompanying inflammatory processes. This review provides some evidence that the positive effects of training are maintained for a period after termination of formal physical training. It is not clear how long these benefits will be maintained for, although it has been suggested that the benefits of training are lost at a rate of one per cent exercise capacity per day 13

17 after stopping training for seven days (McArdle 2001). There are insufficient studies in this review to compare the effectiveness of aerobic versus anaerobic training versus a combination of both of these. A U T H O R S C O N C L U S I O N S Implications for practice Conclusions about the efficacy of physical training in CF are limited by the small size, duration and incomplete reporting of most of the studies included in this review. However, there is limited evidence that physical training is beneficial. The benefits obtained from including physical training in a package of care may be influenced by the type of training programme and the inclusion of aerobic and anaerobic training. Physical training is already part of the care package offered to most people with CF and there is no evidence to actively discourage this. Implications for research Further research is needed to assess comprehensively the benefits of exercise programmes in people with cystic fibrosis and the relative benefits of the addition of aerobic versus anaerobic versus a combination of both types of physical training to the care of people with cystic fibrosis. R E F E R E N C E S References to studies included in this review Cerny 1989 {published data only} Cerny FJ. Relative effects of bronchial drainage and exercise for in-hospital care of patients with cystic fibrosis. Physical Therapy 1989;69(8): Klijn 2004 {published data only} Klijn PH, Oudshoorn A, van der Ent CK, ann der Net J, Helders PJ. Effects of anaerobic training in children with cystic fibrosis: a randomised controlled study. Chest 2004; 125(4): Michel 1989 {published data only} Michel SH, Darbee JC, Pequignot E. Exercise, body composition and strength in cystic fibrosis [abstract]. Pediatric Pulmonology 1989;Suppl 4:116. Moorcroft 2004 {published data only} Dodd ME, Moorcroft AJ, Webb AK. Improved fitness and decreased symptoms of muscle fatigue for upper and lower body following an individualised unsupervised home exercise training programme in adults with cystic fibrosis [abstract]. Pediatric Pulmonology 1998;Suppl 17:347. Moorcroft AJ, Dodd ME, Morris J, Webb AK. Individualised unsupervised exercise training in adults with cystic fibrosis: a 1 year randomised controlled trial. Thorax 2004;59(12): Moorcroft AJ, Dodd ME, Webb AK. Individualised home exercise training in CF - a one- year trial [abstract]. Proceedings of the XIIIth International Cystic Fibrosis Congress; 2000 June 4-8; Stockholm, Sweden. 2000:153. Schneiderman 2000 {published data only} Reisman JJ, Schneiderman-Walker J, Corey M, Wilkes D, Pedder L, Levison H, et al.the role of an organised exercise program in CF - a three year study [abstract]. Pediatric Pulmonology 1995;Suppl 12:261. Schneiderman-Walker J, Pollock SL, Corey M, Wilkes DD, Canny G, Pedder L, et al.a randomised controlled trial of a 3-year home exercise program in cystic fibrosis. Journal of Pediatrics 2000;136(3): Selvadurai 2002 {published data only} Selvadurai HC, Blimkie CJ, Meyers N, MellisCM, Cooper PJ, Van Asperen PP. Randomized controlled study of inhospital exercise training programs in children with cystic fibrosis. Pediatric Pulmonology 2002;33(3): Selvadurai HC, Van Asperen PP, Cooper PJ, Mellis CM, Blimkie CJ. A randomised controlled study of in-hospital exercise training programs in children with cystic fibrosis (CF) [abstract]. Pediatric Pulmonology 1999;Suppl 19: Turchetta 1991 {published data only} Turchetta A, Bella S, Calzolari A, Castro M, Ciuffetti C, Drago F, et al.effect of controlled physical activity on lung function test of cystic fibrosis children [abstract]. Proceedings of the 17th European Cystic Fibrosis Conference; 1991 June 18-21; Copenhagen. 1991:134. References to studies excluded from this review Albinni 2004 {published data only} Albinni S, Rath R, Renner S, Eichler I. Additional inspiratory muscle training intensifies the beneficial effects of cycle ergometer training in patients with cystic fibrosis [abstract]. Journal of Cystic Fibrosis 2004;3(Suppl 1):S63. Eichler I, Renner S, Albinni S, Nachbaur E, Rath R. Inspiratory muscle training adds beneficial effects to cycle ergometer training in patients with cystic fibrosis [abstract]. Pediatric Pulmonology 2005;40(Suppl 28):320. Andreasson 1987 {published data only} Andreasson B, Jonson B, Kornfalt R, Nordmark E, Sandstrom S. Long-term effects of physical exercise on working capacity and pulmonary function in cystic fibrosis. Acta Paediatrica Scandinavica 1987;76(1):70 5. Aquino 2006 {published data only} Aquino A, Balestri E, Dall Ara S, Lami I, Gobb F, Ambroni M, Miano A. Efficacy of physical exercise playing a video game for mucus clearance in patients with CF. Jounal of Cystic Fibrosis 2006;5(Suppl):S83. 14

18 Balestri 2004 {published data only} Balestri E, Ambroni M, dall Ara S, Miano A. Efficacy of physical exercise for mucus clearance in patients with cystic fibrosis (CF). Pediatric Pulmonology 2004;38(S27):316. Barry 2001 {published data only} Barry S, Dodd J, Jensma M, Gallagher C. Benefits of high intensity strength training in adults with cystic fibrosis [abstract]. American Journal of Respiratory and Critical Care Medicine 2001;163(5 Suppl):A968. Barry S, Dodd J, Jensma M, Gallagher C. High intensity strength training improves fitness levels in adults with cystic fibrosis [abstract]. American Journal of Respiratory and Critical Care Medicine 2002;165(8 Suppl):A507. Bilton 1992 {published data only} Bilton D, Dodd M, Webb AK. The benefits of exercise combined with physiotherapy in cystic fibrosis [abstract]. Pediatric Pulmonology 1990;9(Suppl 5):238. Bilton D, Dodd ME, Abbot JV, Webb AK. The benefits of exercise combined with physiotherapy in the treatment of adults with cystic fibrosis. Respiratory Medicine 1992;86 (6): Chatham 1997 {published data only} Chatham K, Ionescu A, Davis C, Baldwin J, Enright S, Shale DJ. Through range computer generated inspiratory muscle training in cystic fibrosis [abstract]. Pediatric Pulmonology 1997;23(Suppl 14):299. de Jong 1994 {published data only} de Jong W, Grevink RG, Roorda RJ, Kaptein AA, van der Schans CP. Effect of a home exercise training program in patients with cystic fibrosis. Chest 1994;105(2): Edlund 1986 {published data only} Adams TD, Edland LL, French RW, Herbst JJ, Ruttenberg HD, Ruhling RO. Effects of a swimming program on children with cystic fibrosis [abstract]. International Journal of Sports Medicine 1984;5:156. Edlund LD, French RW, Herbst JJ, Ruttenburg HD, Ruhling RO, Adams TD. Effects of a swimming program on children with cystic fibrosis. American Journal of Diseases of Children 1986;140(1):80 3. Falk 1988 {published data only} Falk M, Kelstrup M, Andersen JB, Pedersen SS, Rossing I, Dirksen H. PEP treatment or physical exercise - Effects on secretions expectorated and indices of central and peripheral airway function. A controlled study [abstract]. Excerpta Medica, Asia Pacific Congress Series 1988;74:35. Hall 2010 {published data only} Hall K, Peasey M, Wood M, Cobb R, Bell SC, Kuys S. The effects of Nintendo Wii exercise training in adults with cystic fibrosis. Journal of Cystic Fibrosis 2010;9(Suppl 1): A275. Hebestreit 2003 {published data only} Hebestreit A, Kriemler S, Kieser S, Junge S, Heilmann M, Ballmann M, et al.effects of different conditioning programmes on aerobic capacity in CF. Pediatric Pulmonology 2003;36(Suppl 25):329. Heijerman 1992 {published data only} Heijerman HGM, Bakker W, Sterk PJ, Dijkman JH. Longterm effects of exercise training and hyperlimentation in adult cystic fibrosis patients with severe pulmonary dysfunction. International Journal of Rehabilitation Research 1992;15(3): Kriemler 2001 {published data only} Kriemler S, Hebestreit A, Kieser S, Bachmann M, Hebestreit H. Six months of training improves lung function and aerobic performance in CF [abstract]. Abstracts of the 24th European Cystic Fibrosis Conference 6-9 June 2001; Vienna. 2001:P62. Lannefors 1992 {published data only} Lannefors L, Wollmer P. Mucus clearance in cystic fibrosis (CF) - a comparison between postural drainage, PEP-mask and physical exercise [abstract]. Proceedings of the 11th International Cystic Fibrosis Congress. 1992:AHP31. Lannefors L, Wollmerr P. Mucus clearance with three chest physiotherapy regimes in CF: a comparison between postural drainage, PEP and physical exercise. European Respiratory Journal 1992;5(6): Oreinstein 2004 {published data only} Orenstein DM, Hovell MF, Mulvihill M, Keating KK, Hofstetter CR, Kelsey S, et al.strength vs aerobic training in children with cystic fibrosis: a randomised controlled trial. Chest 2004;126(4): Orenstein 1981 {published data only} Orenstein DM, Franklin BA, Doershuk CF, Hellerstein HK, Germann KJ, Horowitz JG, et al.exercise conditioning and cardiopulmonary fitness in CF. The effects of a threemonth supervised running program. Chest 1981;80(4): Salh 1989 {published data only} Salh W, Bilton D, Dodd M, Webb AK. Effect of exercise and physiotherapy in aiding sputum expectoration in adults with cystic fibrosis. Thorax 1989;44(12): Stanghelle 1998 {published data only} Stanghelle JK, Hjeltnes N, Bangstad HJ, Michalsen H. EFfect of short bouts of trampoline exercise during 8 weeks on the pulmonary function and the maximal oxygen uptake of children with cystic fibrosis. International Journal of Sports Medicine 1988;9:3 36. Tuzin 1998 {published data only} Tuzin BJ, Mulvihill MM, Kilbourn KM, Bertran DA, Buono M, Hovell MF, et al.increasing physical activity of children with cystic fibrosis: A home based family intervention. Pediatric Exercise Science 1998;10(1): References to ongoing studies Phillips 2008 {published data only} Phillips AL, Lee L, Britton LJ, Hoover W, Lowman JD. Efficacy of a standardised exercise protocol in inpatient care of patients with cystic fibrosis [abstract]. Pediatric Pulmonology 2008;43(Suppl 31):385, Abstract

19 Additional references ACSM 2010 American College of Sports Medicine. ACSM s guidelines for exercise testing prescription. 8th Edition. Philadelphia: Lippincott Williams and Williams, Casaburi 1992 Casaburi R. Principles of exercise training. Chest 1992;101 (Suppl 5):263S 7S. CF Foundation 2009 Cystic Fibrosis Foundation. Patient Registry: Annual Data Report. Patient Registry: Annual Data Report. [Bethesda, Maryland], CF Trust 2010 Cystic Fibrosis Trust. Living with Cystic Fibrosis: Annual Review. Cystic Fibrosis Trust, Higgins 2003 Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327 (7414): Higgins 2009 Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version [updated September 2009]. The Cochrane Collaboration, Available from Lloyd 2010 Lloyd FA, Collins A, Morris A, Spencer E, Ledson M, Walshaw M. Provision of structured exercise in specialist UK cystic fibrosis centres. Journal of Cystic Fibrosis 2010;9 (Suppl 1):A274. Mazzeo 1998 Mazzeo, et al.harries M, King J, Williams C (ed). ABC of Sports Medicine. London: BMJ Publication, McArdle 2001 McArdle WD, Katch F, Katch V. Exercise physiology, energy nutrition and human performance. Lippincott Williams and Wilkins, Baltimore, Nixon 1992 Nixon PA, Orenstein DM, Kelsey SF, Doerschuk CF. The prognostic value of exercise testing in patients with cystic fibrosis. New England Journal of Medicine 1992;327(25): O Neill 1987 O Neill PA, Dodds M, Phillips B, Poole J, Webb AK. Regular exercise and reduction of breathlessness in patients with cystic fibrosis. British Journal of Diseases of the Chest 1987;81(1):62 9. Peebles 1998 Peebles AD. Physiotherapy (Chapter 5). In: Hill CM editor (s). Practical Guidelines for Cystic Fibrosis Care. 1st Edition. London: Churchill Livingstone, Primos 1996 Primos WA. Sports exercise during acute illness: recommending the right course for the patient. The Physician and Sports Medicine 1996;24(1):44. Review Manager 2008 The Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan) Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, Shephard 1994 Shephard RJ. Aerobic Fitness and Health. Leeds, England: Human Kinetic Publishers, Wolman 1995 Wolman RL. Osteoporosis and exercise. In: McLatchie G, Harris M, King J, Williams C editor(s). ABC of Sports Medicine. 4th Edition. BMJ Publishing, Indicates the major publication for the study 16

20 C H A R A C T E R I S T I C S O F S T U D I E S Characteristics of included studies [ordered by study ID] Cerny 1989 Methods Participants Interventions RCT. Parallel design. During hospital admission for acute exacerbation. 17 participants with CF. Group demographics: mean (SD) age 15.4 years (4.9 years) for exercise group (n = 9 participants). (SD) age 15.9 years (4.9 years) for bronchial hygiene group (n = 8 participants) Short-term aerobic study. Comparison of exercise (2 cycle ergometer sessions) and 1 bronchial hygiene per day during admission: mean (SD) 13 days (3 days) versus 3 bronchial hygiene sessions per day during admission: mean (SD) 13 days (2.6 days) Outcomes Included in this study were: pulmonary function (FVC, ERV, IC, FEV 1, FEF 25 75, RV, FRC, TLC, Raw, SGAW, SaO2, and PFS); and exercise performance during cycle ergometry with load increased by 0.3 W/kg every 2 minutes until participant could continue no longer (SaO2, peak load, EMG activity, peak HR, peak VE to peak load ratio, peak HR to peak load ratio). Coughs (15 minutes post treatment session). Sputum wet and dry weight, sputum volume Notes Risk of bias Bias Authors judgement Support for judgement Adequate sequence generation? Unclear risk Described as randomised but no details of the method. Allocation concealment? Unclear risk Not discussed. Blinding? All outcomes Incomplete outcome data addressed? All outcomes Unclear risk Low risk Not possible to blind participants or clinicians to intervention, unclear whether outcome assessors blinded There were no dropouts. Free of selective reporting? Low risk All outcome detailed in methods were reported in results. Data reported for all timepoints 17

21 Cerny 1989 (Continued) Free of other bias? Unclear risk Stated the inclusion criteria but not the exclusion criteria Clearly described statistical analysis methods. Klijn 2004 Methods Participants Interventions Outcomes Notes RCT. 3-month parallel design. Assigned to group by concealed opaque envelopes. Stable disease. 20 participants with CF completed the study. Group demographics: mean (SD) age 13.6 years (1.3 years) for training group; 14.2 years (2.1 years) for control group. 3 participants dropped out: 2 as they did not complete training due to pulmonary exacerbations; and 1 withdrew for practical reasons Long-term anaerobic study. Comparison of anaerobic exercise (2 days per week for minutes for 12 weeks) versus normal daily activities Included in this study were: BMI; FEV 1 ; FVC; FEF ; RV; TLC; Wingate anaerobic test - VO2max; VCO2; VE; RER; lactate; habitual activity estimation scale; CF questionnaire; fat-free mass; total maximal muscle force. Outcomes measured again at 12 weeks follow up To achieve a difference in PP per kg BW of 10% with an SD of 0.8 W/kg and a statistical power of 805, it was calculated that 8 participants had to be included in each study group Risk of bias Bias Authors judgement Support for judgement Adequate sequence generation? Unclear risk Described as randomised no details of the method. Allocation concealment? Low risk Concealed in opaque envelopes. Blinding? All outcomes Incomplete outcome data addressed? All outcomes Unclear risk Low risk Not possible to blind participants or clinicians to intervention, unclear whether outcome assessors blinded Clear description and details about dropouts: 3 participants dropped out: 2 as they did not complete training due to pulmonary exacerbations; and 1 withdrew for practical 18

22 Klijn 2004 (Continued) reasons Free of selective reporting? Low risk All outcome detailed in methods were reported in results. Data reported for all time points Free of other bias? Low risk Clearly stated inclusion and exclusion criteria and described statistical methods used in analysis Michel 1989 Methods Participants Interventions Outcomes RCT. Parallel design. During hospital admission. 9 participants with CF. Group demographics: mean (SD) age 25.5 years (10.5 years) for exercise group. (SD) age 21.5 years (3.2 years) for non-exercise group Short-term aerobic study. Comparison of exercise and standardised CF protocol versus standardised CF protocol at 1 month post-discharge Included in this study were: skin folds; mid-arm circumference; grip strength; respiratory muscle strength (MSIP and MSEP); ideal body weight Notes Risk of bias Bias Authors judgement Support for judgement Adequate sequence generation? Unclear risk Described as randomised, but no details of method. Allocation concealment? Unclear risk Not discussed. Blinding? All outcomes Incomplete outcome data addressed? All outcomes Unclear risk Unclear risk Not possible to blind participants or clinicians to intervention, unclear whether outcome assessors blinded No details of dropouts or whether intention-to-treat analysis had been used Free of selective reporting? Unclear risk This is an abstract so unable to assess if all outcome used in methods were reported in results. Unable to assess if data were re- 19

23 Michel 1989 (Continued) ported for all time-points Free of other bias? Unclear risk Do not state inclusion or exclusion criteria, nor do they describe the methods of statistical analysis used Moorcroft 2004 Methods Participants Interventions Outcomes Notes RCT. 1-year parallel design. Validity: allocation concealment unclear. 42 participants with CF completed the study. Group demographics: mean (SD) age 23.5 years (6.4 years) for training group; 23.6 years (5.5 years) for control group. 3 participants dropped out at the start of programme: 1 from training group due to failure to attend on initial assessment; and 2 in the control group were withdrawn due to ill health. A further 6 participants dropped out during the 1-year period Long-term aerobic and anaerobic study. Comparison of unsupervised exercise (based on individual preferences general aerobic exercises for lower body and weight training for upper body) 3 times per week over 1 year versus control (continue with usual activities) Included in this study were: whole blood lactate; RER; heart rate; Borg breathlessness and muscle effort; VE, RR peak for arm and bicycle ergometry at 55% maximal workload; and weight This study is a full text article of Dodd 1998 and Moorcroft 2000 abstracts Risk of bias Bias Authors judgement Support for judgement Adequate sequence generation? Low risk Randomised to either active or control groups in a ratio of 3:2. A stratified randomisation in blocks was used to balance the groups for FEV 1, sputum colonisation by Burkholderia cepacia and sex Allocation concealment? Unclear risk Not discussed. Blinding? All outcomes Unclear risk Not possible to blind participants or clinicians to intervention, unclear whether outcome assessors blinded 20

24 Moorcroft 2004 (Continued) Incomplete outcome data addressed? All outcomes Low risk 3 participants dropped out at the start of programme: 1 from training group due to failure to attend on initial assessment; and 2 in the control group were withdrawn due to ill health. A further 6 participants dropped out during the 1-year period Free of selective reporting? Low risk All outcome detailed in methods were reported in results. Data reported for all timepoints Free of other bias? Low risk Clearly stated inclusion and exclusion criteria and described method of statistical analysis used Schneiderman 2000 Methods RCT. 3-year parallel design. Computerised randomisation. Participants 65 participants with CF: exercise group (n = 30 participants); control group (n = 35 participants). Group demographics: mean (SD) age 13.4 years (3.9 years) for exercise group. (SD) age 13.3 years (3.6 years) for control group. 7 dropouts. Similar at baseline. Interventions Outcomes Long-term aerobic study. Comparison of exercise (minimum of 20 minutes aerobic activity plus 5 minute warm up and cool down 3 times per week) versus maintained regular activity (control) Included in this study were: FVC; FEV 1 ; FEF ; PEFR; TV; VO2 peak; VCO2; peak exercise heart rate; peak exercise VE; VE peak/mvv; RER; blood pressure; % of ideal weight for height; compliance and sense of well-being; feasibility of exercise; hospital stays and number of days in hospital; chest X-ray; and Schwachman scores Notes Risk of bias Bias Authors judgement Support for judgement Adequate sequence generation? Low risk Computer-generated randomisation sequence. Allocation concealment? Unclear risk Not discussed. 21

25 Schneiderman 2000 (Continued) Blinding? All outcomes Incomplete outcome data addressed? All outcomes Unclear risk Low risk Not possible to blind participants or clinicians to intervention, unclear whether outcome assessors blinded Clear description and details about 7 dropouts. Free of selective reporting? Low risk All outcome detailed in methods were reported in results. Data reported for all timepoints Free of other bias? Unclear risk Groups similar at baseline. Stated the inclusion criteria but not the exclusion criteria. Described statistical methods used in analysis. Selvadurai 2002 Methods Participants Interventions RCT. Parallel design. Random allocation in sets of 6, using concealed information inside opaque envelopes. Hospital admission for recurrent chest infections. 66 children with CF: aerobic training male: female 9:13; resistance training male: female 10:12; control male: female 9:13. Group demographics: mean (SD) age 13.2 years (2.0 years) for aerobic training. (SD) age 13.1 years (2.1 years) for resistance training group. (SD) age 13.2 years (2.0 years) for control group. No dropouts. Short-term aerobic and anaerobic study. Comparison of aerobic exercise (30 minutes supervised training 5 times per week) versus resistance training (30 minutes supervised training 5 times per week) versus no specific training during hospital admission (mean duration 18.7 days, range days) Outcomes Included in this study were: VO2, peak, VE, VCO2; peak HR; quality of life; FEV 1, FVC; weight; lower limb strength; and fat-free mass. Notes Risk of bias Bias Authors judgement Support for judgement Adequate sequence generation? Low risk Random allocation in sets of 6. 22

26 Selvadurai 2002 (Continued) Allocation concealment? Low risk Concealed information inside opaque envelopes. Blinding? All outcomes Incomplete outcome data addressed? All outcomes Unclear risk Low risk Not possible to blind participants or clinicians to intervention, unclear whether outcome assessors blinded Stated no dropouts. Free of selective reporting? High risk Did not report on all outcomes detailed in methods (e.g. Ve,VCO2 RQ) in results. Data reported for all time-points Free of other bias? Low risk Clearly stated inclusion and exclusion criteria. Described statistical methods used to analyse data. Turchetta 1991 Methods Participants Interventions Outcomes RCT. Parallel design. Validity allocation unclear. Hospital admission for routine assessment of clinical condition 12 children with CF, 8 males, mean age 12.3 years. No group demographics available. Short-term aerobic study. Comparison of exercise (20 minutes running or treadmill per day for 2 weeks) versus normal hospital treatment Included in this study were: FEV 1 and FVC. Notes Risk of bias Bias Authors judgement Support for judgement Adequate sequence generation? Unclear risk Described as randomised, but no details given. Allocation concealment? Unclear risk Not discussed. 23

27 Turchetta 1991 (Continued) Blinding? All outcomes Incomplete outcome data addressed? All outcomes Unclear risk Unclear risk Not possible to blind participants or clinicians to intervention, unclear whether outcome assessors blinded No details of dropouts or whether intention-to-treat analysis had been used Free of selective reporting? Unclear risk This is an abstract so unable to assess if all outcome used in methods were reported in results. Data were reported for all timepoints Free of other bias? Unclear risk Do not state inclusion or exclusion criteria, nor do they describe the methods of statistical analysis used BW: body weight BMI: body mass index CF: cystic fibrosis FEF : forced expiratory flow 25-75% FEV 1 : forced expiratory volume at one second FRC: functional residual capacity FVC: forced vital capacity MSEP: maximal static expiratory mouth pressure MSIP: maximal static inspiratory mouth pressure MVV: maximal voluntary ventilation Nm: Newton metre PFS: progression-free survival PP: peak power Raw: airways resistance RCT: randomised controlled trial; RER: respiratory exchange ratio RR: respiratory rate RV: residual volume SD: standard deviation SGAW: specific airways conductance SAO2: arterial oxyhaemoglobin saturation SEM: standard error of the mean TLC: total lung capacity VE: minute ventilation VO2: oxygen uptake W: watt 24

28 Characteristics of excluded studies [ordered by study ID] Study Albinni 2004 Andreasson 1987 Aquino 2006 Balestri 2004 Barry 2001 Bilton 1992 Chatham 1997 de Jong 1994 Edlund 1986 Falk 1988 Hall 2010 Hebestreit 2003 Heijerman 1992 Kriemler 2001 Lannefors 1992 Oreinstein 2004 Orenstein 1981 Reason for exclusion This study was designed with the exercise group as the control group therefore we could not compare data with baseline, no physical training as per our protocol Not a randomised controlled study. This study was designed with the aim of comparing the effectiveness of a single treatment sessions of exercise and PEP on sputum clearance. Participants in this study did not undertake a programme of physical training This study was designed with the aim of comparing the effectiveness of a single treatment session of exercise and PEP on sputum clearance. Participants in this trial did not undertake a programme of physical training Not a randomised controlled study. This study was designed with the aim of comparing the effectiveness of a single treatment session of exercise or physiotherapy or exercise and physiotherapy on sputum clearance and lung function. Participants in this trial did not undertake a programme of physical training This study involved respiratory muscle training exclusively. This intervention does not constitute physical training as defined within our protocol Not a randomised controlled study. Not a randomised controlled study. This study was designed with the aim of comparing the effectiveness of a single treatment session of exercise or positive expiratory pressure on lung function. Participants in this study did not undertake a programme of physical training Compares Nintendo Wii exercise training to an existing exercise programme, no control group with no physical training Not a randomised controlled study. Not a randomised controlled study. Not a randomised controlled study. This study was designed with the aim of comparing the effectiveness of a single treatment session of exercise and FET or positive expiratory pressure and FET or postural drainage, thoracic expansion exercises and FET on mucous clearance. Participants in this study did not undertake programme of physical training Compares aerobic training to upper-body strength training, no control group with no physical training Not a randomised controlled study. 25

29 (Continued) Salh 1989 Stanghelle 1998 Tuzin 1998 Not a randomised controlled study. Not a randomised controlled study. Not a randomised controlled study. FET: forced expiration technique Characteristics of ongoing studies [ordered by study ID] Phillips 2008 Trial name or title Methods Participants Interventions Outcomes Randomised, parallel trial. 10 patients (of an anticipated 30), age 6-21, with an FEV < 60%, were admitted for a 10 day hospitalization for CF exacerbation Standardized moderate-to-high intensity RT and AT exercise program compared to AT alone for PT management of a CF exacerbation during an inpatient hospital stay The experimental group received 1 hour of RT and flexibility training 3 days/week and minutes of AT and balance training 2 days/week The control group received the current standard of care which included minutes of variable intensity AT 5 days/week MST and multiple measures of peripheral muscle performance at admission and discharge. Adverse effects Starting date Contact information Notes Some results presented, but abstract states still recruiting AT: aerobic training CF: cystic fibrosis~ FEV: forced expiratory volume MST: modified shuttle test PT: physical therapy RT: resistive training 26

30 D A T A A N D A N A L Y S E S Comparison 1. Aerobic training versus no physical training Outcome or subgroup title No. of studies No. of participants Statistical method Effect size 1 Change in VO2 max during 1 (IV, Fixed, 95% CI) Totals not selected maximal exercise test (ml/kg/min) 2 Annual rate change in VO2max 1 (IV, Fixed, 95% CI) Totals not selected during maximal exercise test (ml/kg/min) 3 Annual rate decline peak heart 1 (IV, Fixed, 95% CI) Totals not selected rate (beats/min) 4 Change in saturation during 1 (IV, Fixed, 95% CI) Totals not selected maximal exercise test (%) 5 Annual rate of change in VE 1 (IV, Fixed, 95% CI) Totals not selected (L/min) 6 Annual rate change in peak 1 (IV, Fixed, 95% CI) Totals not selected working capacity during maximal exercise test (%) 7 Change in FEV1(%) 1 (IV, Fixed, 95% CI) Totals not selected 8 Change in strength (Newton 1 (IV, Fixed, 95% CI) Totals not selected metres) 9 Annual rate of change in FEV1 1 (IV, Fixed, 95% CI) Totals not selected (%) 10 Change FVC (%) 1 (IV, Fixed, 95% CI) Totals not selected 11 Annual rate of change in FVC 1 (IV, Fixed, 95% CI) Totals not selected (%) 12 Annual rate of change in FEF 1 (IV, Fixed, 95% CI) Totals not selected (%) 13 Change in quality of life 1 (IV, Fixed, 95% CI) Totals not selected 14 Change in weight (kg) 1 (IV, Fixed, 95% CI) Totals not selected 15 Change in fat free mass (kg) 1 (IV, Fixed, 95% CI) Totals not selected 16 Annual change of ideal weight 1 (IV, Fixed, 95% CI) Totals not selected for height (%) 17 Change in activity 1 (IV, Fixed, 95% CI) Totals not selected 27

31 Comparison 2. Anaerobic training versus no physical training Outcome or subgroup title No. of studies No. of participants Statistical method Effect size 1 Change in VO2 max 2 (IV, Fixed, 95% CI) Totals not selected (ml/kg/min) 1.1 At hospital discharge 1 (IV, Fixed, 95% CI) 0.0 [0.0, 0.0] 1.2 At 3 months 1 (IV, Fixed, 95% CI) 0.0 [0.0, 0.0] 2 Change in lactate during 1 (IV, Fixed, 95% CI) Totals not selected maximal test (mmol/l) 3 Change in saturation during 1 (IV, Fixed, 95% CI) Totals not selected maximal exercise test (%) 4 Change in peak power during 1 (IV, Fixed, 95% CI) Totals not selected maximal test (watts) 5 Change in mean power during 1 (IV, Fixed, 95% CI) Totals not selected maximal test (watts) 6 Change in maximum workload 1 (IV, Fixed, 95% CI) Totals not selected during maximal test (watts) 7 Change in lower limb strength 1 (IV, Fixed, 95% CI) Totals not selected (Newton metres) 8 Change in FEV1 (%) 1 (IV, Fixed, 95% CI) Totals not selected 9 Change in FVC (%) 1 (IV, Fixed, 95% CI) Totals not selected 10 Change in physical function 1 (IV, Fixed, 95% CI) Totals not selected (CF questionnaire) 11 Change in quality of life 1 (IV, Fixed, 95% CI) Totals not selected 12 Change in weight (kg) 1 (IV, Fixed, 95% CI) Totals not selected 13 Change in fat free mass (kg) 1 (IV, Fixed, 95% CI) Totals not selected 14 Change in activity 1 (IV, Fixed, 95% CI) Totals not selected Comparison 3. Combined aerobic and anaerobic training Outcome or subgroup title No. of studies No. of participants Statistical method Effect size 1 Annual change in peak heart rate during constant load bicycle ergometry (beat/min) 2 Annual change in peak heart rate during constant load arm ergometry (beat/min) 3 Annual change in Ve (L/min) during constant load bicycle ergometry 4 Annual change in Ve (L/min) during constant load arm ergometry 1 Heart rate (Fixed, 95% CI) Totals not selected 1 peak heart rate (Fixed, 95% CI) Totals not selected 1 Ve (Fixed, 95% CI) Totals not selected 1 Ve (Fixed, 95% CI) Totals not selected 28

32 5 Annual change in lactate 1 Lactate (Fixed, 95% CI) Totals not selected (mmol/l)during constant load bicycle ergometry 6 Annual change in lactate 1 Lactate (Fixed, 95% CI) Totals not selected (mmol/l) during constant load arm ergometry (beat/min) 7 Annual change in RER during 1 RER (Fixed, 95% CI) Totals not selected constant load bicycle ergometry 8 Annual change in RER during 1 RER (Fixed, 95% CI) Totals not selected constant load arm ergometry 9 Annual change in RR during 1 RR (Fixed, 95% CI) Totals not selected constant load bicycle ergometry (breaths/min) 10 Annual change in RR during 1 RR (Fixed, 95% CI) Totals not selected constant load arm ergometry (breaths/min) 11 Annual change in Borg 1 Borg breathlessness (Fixed, 95% CI) Totals not selected breathlessness during constant load bicycle ergometry 12 Annual change in Borg 1 Borg breathlessness (Fixed, 95% CI) Totals not selected breathlessness during constant load arm ergometry 13 Annual change in Borg muscle 1 Borg muscle effort (Fixed, 95% CI) Totals not selected effort during constant load bicycle ergometry 14 Annual change in Borg muscle 1 Borg muscle effort (Fixed, 95% CI) Totals not selected effort during constant load arm ergometry 15 Annual change in FEV1 (ml) 1 Annual change in FEV (Fixed, 95% CI) Totals not selected 16 Annual change in FVC(ml) 1 Annual change in FVC (Fixed, 95% CI) Totals not selected 17 Annual change in BMI (kg/ml) 1 Annual change in BMI (Fixed, 95% CI) Totals not selected Analysis 1.1. Comparison 1 Aerobic training versus no physical training, Outcome 1 Change in VO2 max during maximal exercise test (ml/kg/min). 1 Aerobic training versus no physical training 1 Change in VO2 max during maximal exercise test (ml/kg/min) Study or subgroup Aerobic training Control N (SD) N (SD) Selvadurai (6.29) (6.15) 8.53 [ 4.85, ] Favours AT 29

33 Analysis 1.2. Comparison 1 Aerobic training versus no physical training, Outcome 2 Annual rate change in VO2max during maximal exercise test (ml/kg/min). 1 Aerobic training versus no physical training 2 Annual rate change in VO2max during maximal exercise test (ml/kg/min) Study or subgroup Aerobic training Control N (SD) N (SD) Schneiderman (2.21) (2.51) 0.05 [ -1.10, 1.20 ] Favours AT Analysis 1.3. Comparison 1 Aerobic training versus no physical training, Outcome 3 Annual rate decline peak heart rate (beats/min). 1 Aerobic training versus no physical training 3 Annual rate decline peak heart rate (beats/min) Study or subgroup Favours AT N (SD) N (SD) Schneiderman (3.68) (4.33) 1.10 [ -0.85, 3.05 ] Favours treatment 30

34 Analysis 1.4. Comparison 1 Aerobic training versus no physical training, Outcome 4 Change in saturation during maximal exercise test (%). 1 Aerobic training versus no physical training 4 Change in saturation during maximal exercise test (%) Study or subgroup Aerobic training Control N (SD) N (SD) Selvadurai (0.43) (0.56) 0.62 [ 0.32, 0.92 ] Favours AT Analysis 1.5. Comparison 1 Aerobic training versus no physical training, Outcome 5 Annual rate of change in VE (L/min). 1 Aerobic training versus no physical training 5 Annual rate of change in VE (L/min) Study or subgroup Favours AT Favours Control N (SD) N (SD) Schneiderman (8.31) (6.57) 2.09 [ -1.60, 5.78 ] Favours AT 31

35 Analysis 1.6. Comparison 1 Aerobic training versus no physical training, Outcome 6 Annual rate change in peak working capacity during maximal exercise test (%). 1 Aerobic training versus no physical training 6 Annual rate change in peak working capacity during maximal exercise test (%) Study or subgroup Aerobic training Control N (SD) N (SD) Schneiderman (5.16) (6.08) 0.82 [ -1.91, 3.55 ] Favours AT Analysis 1.7. Comparison 1 Aerobic training versus no physical training, Outcome 7 Change in FEV1(%). 1 Aerobic training versus no physical training 7 Change in FEV1(%) Study or subgroup Treatment Control N (SD) N (SD) Selvadurai (7.76) (6.9) 2.03 [ -2.31, 6.37 ] Favours AT 32

36 Analysis 1.8. Comparison 1 Aerobic training versus no physical training, Outcome 8 Change in strength (Newton metres). 1 Aerobic training versus no physical training 8 Change in strength (Newton metres) Study or subgroup Aerobic training Control N (SD) N (SD) Selvadurai (6.23) (6.1) 8.13 [ 4.49, ] Favours AT Analysis 1.9. Comparison 1 Aerobic training versus no physical training, Outcome 9 Annual rate of change in FEV1 (%). 1 Aerobic training versus no physical training 9 Annual rate of change in FEV1 (%) Study or subgroup Aerobic training Control N (SD) N (SD) Schneiderman (3.55) (4.93) 2.01 [ -0.06, 4.08 ] Favours AT 33

37 Analysis Comparison 1 Aerobic training versus no physical training, Outcome 10 Change FVC (%). 1 Aerobic training versus no physical training 10 Change FVC (%) Study or subgroup Treatment Control N (SD) N (SD) Selvadurai (4.62) (4.22) 0.06 [ -2.55, 2.67 ] Favours AT Analysis Comparison 1 Aerobic training versus no physical training, Outcome 11 Annual rate of change in FVC (%). 1 Aerobic training versus no physical training 11 Annual rate of change in FVC (%) Study or subgroup Aerobic training Control N (SD) N (SD) Schneiderman (2.81) (4.15) 2.17 [ 0.47, 3.87 ] Favours AT 34

38 Analysis Comparison 1 Aerobic training versus no physical training, Outcome 12 Annual rate of change in FEF (%). 1 Aerobic training versus no physical training 12 Annual rate of change in FEF (%) Study or subgroup Aerobic training Control N (SD) N (SD) Schneiderman (5.34) (7) 0.80 [ -2.20, 3.80 ] Favours AT Analysis Comparison 1 Aerobic training versus no physical training, Outcome 13 Change in quality of life. 1 Aerobic training versus no physical training 13 Change in quality of life Study or subgroup Aerobic training Control N (SD) N (SD) Selvadurai (0.12) (0.12) 0.10 [ 0.03, 0.17 ] Favours AT 35

39 Analysis Comparison 1 Aerobic training versus no physical training, Outcome 14 Change in weight (kg). 1 Aerobic training versus no physical training 14 Change in weight (kg) Study or subgroup Aerobic training Control N (SD) N (SD) Selvadurai (0.64) (0.58) [ -0.59, 0.13 ] Favours AT Analysis Comparison 1 Aerobic training versus no physical training, Outcome 15 Change in fat free mass (kg). 1 Aerobic training versus no physical training 15 Change in fat free mass (kg) Study or subgroup Aerobic training Control N (SD) N (SD) Selvadurai (0.37) (0.32) 0.01 [ -0.19, 0.21 ] Favours AT 36

40 Analysis Comparison 1 Aerobic training versus no physical training, Outcome 16 Annual change of ideal weight for height (%). 1 Aerobic training versus no physical training 16 Annual change of ideal weight for height (%) Study or subgroup Aerobic training Control N (SD) N (SD) Schneiderman (2.52) (2.75) 0.52 [ -0.76, 1.80 ] Favours AT Analysis Comparison 1 Aerobic training versus no physical training, Outcome 17 Change in activity. 1 Aerobic training versus no physical training 17 Change in activity Study or subgroup Aerobic training Control N (SD) N (SD) Selvadurai (2.44) (2.29) 1.20 [ -0.20, 2.60 ] Favours AT 37

41 Analysis 2.1. Comparison 2 Anaerobic training versus no physical training, Outcome 1 Change in VO2 max (ml/kg/min). 2 Anaerobic training versus no physical training 1 Change in VO2 max (ml/kg/min) Study or subgroup Anaerobic training Control 1 At hospital discharge N (SD) N (SD) Selvadurai (5.89) (6.15) 1.95 [ -1.61, 5.51 ] 2 At 3 months Klijn (2.6) (1.9) 2.10 [ 0.12, 4.08 ] Favours ANT Analysis 2.2. Comparison 2 Anaerobic training versus no physical training, Outcome 2 Change in lactate during maximal test (mmol/l). 2 Anaerobic training versus no physical training 2 Change in lactate during maximal test (mmol/l) Study or subgroup Anaerobic training Control N (SD) N (SD) Klijn (1.4) (2.9) 3.40 [ 1.33, 5.47 ] Favours ANT 38

42 Analysis 2.3. Comparison 2 Anaerobic training versus no physical training, Outcome 3 Change in saturation during maximal exercise test (%). 2 Anaerobic training versus no physical training 3 Change in saturation during maximal exercise test (%) Study or subgroup Anaerobic training Control N (SD) N (SD) Selvadurai (0.4) (0.56) 0.33 [ 0.04, 0.62 ] Favours ANT Analysis 2.4. Comparison 2 Anaerobic training versus no physical training, Outcome 4 Change in peak power during maximal test (watts). 2 Anaerobic training versus no physical training 4 Change in peak power during maximal test (watts) Study or subgroup Anaerobic training Control N (SD) N (SD) Klijn (23.8) (53.7) [ 32.50, ] Favours ANT 39

43 Analysis 2.5. Comparison 2 Anaerobic training versus no physical training, Outcome 5 Change in mean power during maximal test (watts). 2 Anaerobic training versus no physical training 5 Change in mean power during maximal test (watts) Study or subgroup Anaerobic training Control N (SD) N (SD) Klijn (11.8) (29.9) [ 22.56, ] Favours ANT Analysis 2.6. Comparison 2 Anaerobic training versus no physical training, Outcome 6 Change in maximum workload during maximal test (watts). 2 Anaerobic training versus no physical training 6 Change in maximum workload during maximal test (watts) Study or subgroup Anaerobic training Control N (SD) N (SD) Klijn (14) 9-2 (5) [ 4.11, ] Favours ANT 40

44 Analysis 2.7. Comparison 2 Anaerobic training versus no physical training, Outcome 7 Change in lower limb strength (Newton metres). 2 Anaerobic training versus no physical training 7 Change in lower limb strength (Newton metres) Study or subgroup Anaerobic training Control N (SD) N (SD) Selvadurai (7.02) (6.1) [ 20.73, ] Favours ANT Analysis 2.8. Comparison 2 Anaerobic training versus no physical training, Outcome 8 Change in FEV1 (%). 2 Anaerobic training versus no physical training 8 Change in FEV1 (%) Study or subgroup Anaerobic training Control N (SD) N (SD) Selvadurai (7.43) (6.9) 5.58 [ 1.34, 9.82 ] Favours ANT 41

45 Analysis 2.9. Comparison 2 Anaerobic training versus no physical training, Outcome 9 Change in FVC (%). 2 Anaerobic training versus no physical training 9 Change in FVC (%) Study or subgroup Anaerobic training Control N (SD) N (SD) Selvadurai (4.18) (4.22) 0.17 [ -2.31, 2.65 ] Favours ANT Analysis Comparison 2 Anaerobic training versus no physical training, Outcome 10 Change in physical function (CF questionnaire). 2 Anaerobic training versus no physical training 10 Change in physical function (CF questionnaire) Study or subgroup Anaerobic training Control N (SD) N (SD) Klijn (9) (17.9) 1.30 [ , ] Favours ANT 42

46 Analysis Comparison 2 Anaerobic training versus no physical training, Outcome 11 Change in quality of life. 2 Anaerobic training versus no physical training 11 Change in quality of life Study or subgroup Anaerobic training Control N (SD) N (SD) Selvadurai (0.1) (0.12) 0.03 [ -0.04, 0.10 ] Favours ANT Analysis Comparison 2 Anaerobic training versus no physical training, Outcome 12 Change in weight (kg). 2 Anaerobic training versus no physical training 12 Change in weight (kg) Study or subgroup Anaerobic training Control N (SD) N (SD) Selvadurai (0.7) (0.58) 1.73 [ 1.35, 2.11 ] Favours ANT 43

47 Analysis Comparison 2 Anaerobic training versus no physical training, Outcome 13 Change in fat free mass (kg). 2 Anaerobic training versus no physical training 13 Change in fat free mass (kg) Study or subgroup Anaerobic training Control N (SD) N (SD) Selvadurai (0.46) (0.32) 1.80 [ 1.57, 2.03 ] Favours ANT Analysis Comparison 2 Anaerobic training versus no physical training, Outcome 14 Change in activity. 2 Anaerobic training versus no physical training 14 Change in activity Study or subgroup Anaerobic training Control N (SD) N (SD) Selvadurai (2.2) (2.29) 0.65 [ -0.68, 1.98 ] Favours ANT 44

48 Analysis 3.1. Comparison 3 Combined aerobic and anaerobic training, Outcome 1 Annual change in peak heart rate during constant load bicycle ergometry (beat/min). 3 Combined aerobic and anaerobic training 1 Annual change in peak heart rate during constant load bicycle ergometry (beat/min) Study or subgroup Heart rate (SE) Heart rate Heart rate Moorcroft (3.78) [ , ] Favours CT Favours Control Analysis 3.2. Comparison 3 Combined aerobic and anaerobic training, Outcome 2 Annual change in peak heart rate during constant load arm ergometry (beat/min). 3 Combined aerobic and anaerobic training 2 Annual change in peak heart rate during constant load arm ergometry (beat/min) Study or subgroup peak heart rate (SE) peak heart rate peak heart rate Moorcroft (4.49) [ , 7.40 ] Favours CT 45

49 Analysis 3.3. Comparison 3 Combined aerobic and anaerobic training, Outcome 3 Annual change in Ve (L/min) during constant load bicycle ergometry. 3 Combined aerobic and anaerobic training 3 Annual change in Ve (L/min) during constant load bicycle ergometry Study or subgroup Ve (SE) Ve Ve Moorcroft (1.84) [ -6.11, 1.11 ] Favours CT Favours Control Analysis 3.4. Comparison 3 Combined aerobic and anaerobic training, Outcome 4 Annual change in Ve (L/min) during constant load arm ergometry. 3 Combined aerobic and anaerobic training 4 Annual change in Ve (L/min) during constant load arm ergometry Study or subgroup Ve (SE) Ve Ve Moorcroft (1.58) [ -6.40, ] Favours CT 46

50 Analysis 3.5. Comparison 3 Combined aerobic and anaerobic training, Outcome 5 Annual change in lactate (mmol/l)during constant load bicycle ergometry. 3 Combined aerobic and anaerobic training 5 Annual change in lactate (mmol/l)during constant load bicycle ergometry Study or subgroup Lactate (SE) Lactate Lactate Moorcroft (0.36) [ -1.54, ] Favours CT Analysis 3.6. Comparison 3 Combined aerobic and anaerobic training, Outcome 6 Annual change in lactate (mmol/l) during constant load arm ergometry (beat/min). 3 Combined aerobic and anaerobic training 6 Annual change in lactate (mmol/l) during constant load arm ergometry (beat/min) Study or subgroup Lactate (SE) Lactate Lactate Moorcroft (0.42) [ -1.14, 0.50 ] Favours CT 47

51 Analysis 3.7. Comparison 3 Combined aerobic and anaerobic training, Outcome 7 Annual change in RER during constant load bicycle ergometry. 3 Combined aerobic and anaerobic training 7 Annual change in RER during constant load bicycle ergometry Study or subgroup RER (SE) RER RER Moorcroft (0.02) 0.02 [ -0.02, 0.06 ] Favours CT Analysis 3.8. Comparison 3 Combined aerobic and anaerobic training, Outcome 8 Annual change in RER during constant load arm ergometry. 3 Combined aerobic and anaerobic training 8 Annual change in RER during constant load arm ergometry Study or subgroup RER (SE) RER RER Moorcroft (0.02) 0.0 [ -0.04, 0.04 ] Favours CT 48

52 Analysis 3.9. Comparison 3 Combined aerobic and anaerobic training, Outcome 9 Annual change in RR during constant load bicycle ergometry (breaths/min). 3 Combined aerobic and anaerobic training 9 Annual change in RR during constant load bicycle ergometry (breaths/min) Study or subgroup RR (SE) RR RR Moorcroft (2.09) [ -4.90, 3.30 ] Favours CT Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 10 Annual change in RR during constant load arm ergometry (breaths/min). 3 Combined aerobic and anaerobic training 10 Annual change in RR during constant load arm ergometry (breaths/min) Study or subgroup RR (SE) RR RR Moorcroft (2.35) 1.50 [ -3.11, 6.11 ] Favours CT 49

53 Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 11 Annual change in Borg breathlessness during constant load bicycle ergometry. 3 Combined aerobic and anaerobic training 11 Annual change in Borg breathlessness during constant load bicycle ergometry Study or subgroup Borg breathlessness (SE) Borg breathlessness Borg breathlessness Moorcroft (0.51) 0.0 [ -1.00, 1.00 ] Favours CT Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 12 Annual change in Borg breathlessness during constant load arm ergometry. 3 Combined aerobic and anaerobic training 12 Annual change in Borg breathlessness during constant load arm ergometry Study or subgroup Borg breathlessness (SE) Borg breathlessness Borg breathlessness Moorcroft (0.51) [ -1.90, 0.10 ] Favours CT 50

54 Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 13 Annual change in Borg muscle effort during constant load bicycle ergometry. 3 Combined aerobic and anaerobic training 13 Annual change in Borg muscle effort during constant load bicycle ergometry Study or subgroup Borg muscle effort (SE) Borg muscle effort Borg muscle effort Moorcroft (0.61) [ -1.50, 0.90 ] Favours CT Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 14 Annual change in Borg muscle effort during constant load arm ergometry. 3 Combined aerobic and anaerobic training 14 Annual change in Borg muscle effort during constant load arm ergometry Study or subgroup Borg muscle effort (SE) Borg muscle effort Borg muscle effort Moorcroft (0.66) 0.30 [ -0.99, 1.59 ] Favours CT 51

55 Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 15 Annual change in FEV1 (ml). 3 Combined aerobic and anaerobic training 15 Annual change in FEV1 (ml) Study or subgroup Annual change in FEV (SE) Annual change in FEV Annual change in FEV Moorcroft (92.34) [ , ] Favours CT Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 16 Annual change in FVC(ml). 3 Combined aerobic and anaerobic training 16 Annual change in FVC(ml) Study or subgroup Annual change in FVC (SE) Annual change in FVC Annual change in FVC Moorcroft (107.14) [ 3.01, ] Favours CT 52

56 Analysis Comparison 3 Combined aerobic and anaerobic training, Outcome 17 Annual change in BMI (kg/ml). 3 Combined aerobic and anaerobic training 17 Annual change in BMI (kg/ml) Study or subgroup Annual change in BMI (SE) Annual change in BMI Annual change in BMI Moorcroft (0.32) 0.54 [ -0.09, 1.17 ] Favours CT W H A T S N E W Last assessed as up-to-date: 7 March Date Event Description 22 May 2012 Amended Contact details updated. H I S T O R Y Protocol first published: Issue 4, 2000 Review first published: Issue 2, 2002 Date Event Description 7 March 2011 New search has been performed A total of two new references were identified in a search of the Group s CF Trials Register. One study was excluded as it compared Nintendo Wii exercise training to an existing exercise programme and hence did not meet the inclusion criteria (Hall 2010). The other study did meet the inclusion criteria but outlined in its abstract that recruitment was ongoing and for this reason it has been listed as an ongoing study; results 53

57 (Continued) will be included in the review once the study has been completed (Phillips 2008). In addition some amendments were made to the Background in order to incorporate updated guidelines and a relevant survey 19 January 2009 Amended The fourth primary outcome mortality was moved to Secondary outcomes in line with Cochrane Collaboration guidance to limit the number of primary outcomes to three 5 January 2009 New search has been performed A search of the Group s Cystic Fibrosis Trials Register did not identify any references to trials which are potentially eligible for inclusion in this review 12 November 2008 Amended Converted to new review format. 14 November 2007 New search has been performed The search identified 11 new references. Of these, two were additional references to already excluded studies (Albinni 2004; Edlund 1986). The remaining nine studies did not fulfill the inclusion criteria; four of these studies which seemed eligible from the title, have been excluded on the basis of trial design and are listed under Excluded studies (Acquino 2006; Balestri 2004; Orenstein 2004; Stanghelle 1998). The study which was previously listed as Awaiting assessment has been moved to the list of excluded studies after correspondence with the study authors (Hebestreit 2003) 14 November 2007 Amended The generic inverse variance method has been used to analyse data which were previously not able to be presented in the Statistical Analysis. 13 November 2007 New citation required and conclusions have changed Substantive amendment The Synopsis has been replaced by a new Plain Language Summary 25 May 2005 New search has been performed A further article has been included (Klijn 2004). The full paper of the trial by Moorcroft (Moorcroft 2004) has also been included. Following publication of this paper, the details about the published abstracts of this trial, previously listed in the Characteristics of included studies table, under Dodd 1998 and Moorcroft 2000 have been listed under Moorcroft 2004 We contacted authors of trials already included in the review regarding confirmation of data and requests for 54

58 (Continued) additional data. Their responses have been included in section detailing the search strategy One trial has been moved from the Studies awaiting assessment section to the Excluded studies section of the review (Tuzin 1998) One trial has been added to the section Studies awaiting assessment section (Hebestreit 2003). The authors have been contacted and have indicated that this study is in preparation for publication 31 July 2003 Amended The presentation of the data in MetaView has been reformatted 31 July 2003 New search has been performed The full paper of the Selvadurai trial has now been included, previously only the abstract of this trial was included in the review (Selvadurai 2002) A further two trials added to the Excluded studies section of the review (Barry 2001; Kriemler 2001) C O N T R I B U T I O N S O F A U T H O R S The title for the protocol was conceived by the Cochrane Cystic Fibrosis and Genetic Disorders Group. Both J. Bradley and F. Moran designed and assisted in writing the protocol and the review and the updates. J. Bradley acts as guarantor for this review. D E C L A R A T I O N S O F None known. I N T E R E S T D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W The fourth primary outcome mortality was moved to Secondary outcomes in line with Cochrane Collaboration guidance to limit the number of primary outcomes to three. 55

59 I N D E X T E R M S Medical Subject Headings (MeSH) Exercise Therapy; Cystic Fibrosis [ rehabilitation]; Exercise Tolerance; Randomized Controlled Trials as Topic MeSH check words Humans 56