Intensity and type of exercise for lower limb endurance training to optimise exercise capacity for people with chronic obstructive pulmonary disease
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To determine the effects of training intensity or type (i.e. continuous or interval training) on exercise capacity in people with COPD.
Background
Strong evidence exists that pulmonary rehabilitation (PR) improves exercise capacity, reduces dyspnoea and improves quality of life in patients with chronic obstructive pulmonary disease (COPD) Lacasse 2006. Compared to other components of PR, such as education, nutritional and psychosocial interventions, endurance training of the lower limbs has the strongest level of evidence in reversing exercise intolerance and is, therefore, an important component of PR BTS 2001; Ries 2007. In people with COPD, exercise intolerance often results from a complex interaction between symptoms, ventilatory constraints, gas exchange limitations, cardiac dysfunction, poor nutrition and peripheral muscle fatigue Jones 2000; Nici 2006; Pepin 2007. The multifactorial limitation to exercise makes defining and prescribing optimal exercise intensity challenging. Current guidelines are non‐specific, recommending varying levels and methods of prescribing exercise intensity BTS 2001; Nici 2006; Ries 2007. Different levels of exercise intensity have been shown to elicit varying amounts of physiological changes Butcher 2006, but the exercise intensity that ensures optimal physiological benefits without compromising compliance and adherence remains unclear.
Fatigue and dyspnoea are the main symptoms that limit participation in activities of daily living (ADL) and exercise and have been strongly associated with peripheral muscle dysfunction and dynamic hyperinflation respectively in these patients Jones 2000; Pepin 2007. Following lower limb endurance training, a substantial proportion of patients experience less fatigue and dyspnoea, as a result of improved muscle aerobic capacity and reduced dynamic hyperinflation Casaburi 1991; Casaburi 1997; Maltais 1996a; Nici 2006. The improvement in aerobic capacity has been related to increased capillary density and concentrations of oxidative enzymes allowing greater delivery of oxygen within the trained muscles Maltais 1996b; Man 2003. These changes prevent early onset of lactic acidosis and fatigue at low levels of activity in individuals with COPD Maltais 1996a. While exercise training does not alter lung function, improved aerobic capacity resulting in lower lactate levels, reduces the ventilatory stimulus thus reducing ventilation at equivalent submaximal work rates after training Casaburi 1991; Maltais 1996a. Such reductions in ventilatory requirements reduce dynamic hyperinflation and, consequently, exertional dyspnoea Gigliotti 2003; O'Donnell 2001. Lower limb training has also been shown to induce a slower and deeper breathing pattern following exercise and correlates with an increase in exercise endurance time Casaburi 1997; Puente‐Maestu 2006. The change in pattern of breathing may also partially explain the reduction in dyspnoea. Gains in exercise capacity following exercise have been measured by increased maximal oxygen uptake (VO2max) or peak work rate (Wmax) Casaburi 1991; Gosselink 1997; Maltais 1997; Vogiatzis 1999, longer endurance exercise time Ries 1995; Pitta 2004 and greater six‐minute walk distance (6MWD) Pitta 2004; Troosters 2000.
An appropriate exercise prescription is important to elicit these physiological adaptations. Exercise prescription includes the components of intensity, frequency, duration, type (i.e continuous or interval training) and mode of exercise (for example, cycling or walking). Of all these components, the intensity is the most difficult to prescribe for people with COPD due to the multifactorial limitation to exercise which require prescription of exercise intensity to differ from that of a healthy population. The minimum intensity required to produce clinically meaningful responses (i.e the critical training intensity) has not been identified with certainty ACSM 2006; Cooper 2001; Lacasse 2006; Nici 2006; Troosters 2005. Guidelines variably recommend the minimum intensity at 40% of VO2maxCooper 2001, 50% of VO2maxACSM 2006 or 50‐60% of WmaxBTS 2001; Nici 2006; Troosters 2005; Vogiatzis 1999. High intensity exercise training has been regarded as training in the range of 60‐90% Wmax , however this not been systematically evaluated Gosselink 1997; Nici 2006; Ries 2007; Puhan 2005.
There is consensus that the higher the aerobic training intensity, the greater the physiological training effects Nici 2006; Troosters 2005. The influence of high intensity exercise training on other key patient‐centred outcomes such as quality of life, dyspnoea, general health status and long‐term adherence to exercise program is still questionable due to lack of rigorous investigation Maltais 2008; Ries 2007. It has been shown that low intensity exercise, such as performing callisthenics, elicits similar health‐related quality of life (HRQoL) improvements as higher intensity exercise Normandin 2002. Guidelines also recognise the possible benefits associated with exercising at lower intensity in regards to greater long‐term adherence to exercise training Nici 2006; Ries 2007 and less risk of injury Maltais 1997.
Some authors have reported that high intensity training of the lower limbs for at least 20‐30 minutes continuously was well‐tolerated by patients with COPD of all levels of disease severity Casaburi 1991; Casaburi 1997; Puente‐Maestu 2000; Punzal 1991. However, Maltais and co‐authors found that many patients with moderate and severe disease could not meet the intensity target of continuous exercise at 80% of Wmax for 20‐30 minutes, even after a 12‐week training program Maltais 1997. As an alternative to continuous exercise, interval training has been suggested to allow patients with COPD to exercise for brief periods of 1‐3 minutes at high intensity alternated with short periods of recovery, with the overall period of high intensity exercise kept at 20‐30 minutes Gosselink 1997; Nici 2006; Puhan 2005; Troosters 2005. Such interval training is associated with lower levels of lactate accumulation Nici 2006; Troosters 2005. Although both types of exercise improved exercise capacity, dyspnoea and HRQoL to a similar degree Puhan 2005, this has not been evaluated by meta‐analysis.
Although there have been consensus statements and narrative reviews on exercise prescription for patients with COPD, the influence of training intensity on exercise capacity, functional performance and HRQoL measures have not been evaluated in a systematic review . Since continuous or interval training is often conducted at different levels of intensity, the influence of the type of training also needs to be investigated. To date, no review has performed a meta‐analysis of studies comparing groups performing lower limb endurance exercise at different levels of intensity delivered either during continuous or interval training. Therefore, a systematic review with meta‐analysis is necessary to investigate the optimal exercise training intensity and type of training in COPD.
Objectives
To determine the effects of training intensity or type (i.e. continuous or interval training) on exercise capacity in people with COPD.
Methods
Criteria for considering studies for this review
Types of studies
The review will consider randomised controlled trials.
Types of participants
The review will include randomised controlled trials in which all of the participants were diagnosed with stable COPD defined by best post bronchodilator forced expiratory volume in one second (FEV1) / forced vital capacity (FVC) ratio <0.7 GOLD 2008.
Types of interventions
Trials of lower limb exercise training of four weeks or more will be eligible for inclusion if they compare exercise interventions:
a) of different intensities (with type, duration, frequency and mode of exercise the same)
b) of different types (with intensity, duration, frequency and mode of exercise the same)
Recognising that the total amount of work per session is usually matched between groups by manipulating intensity and duration, trials will also
be eligible for inclusion if they compare exercise interventions:
c) of different intensities (with work per session, type, frequency and mode of exercise the same)
d) of different types (with work per session, frequency and mode of exercise the same).
Studies that only compare exercise training with no exercise training will be excluded.
Types of outcome measures
Outcomes measured immediately post‐intervention will be used in this review . Studies that only measure outcomes at periods following completion of the intervention (to determine retention of training effects) will be excluded.
Primary outcomes
-
Maximal exercise capacity: peak VO2 or peak work rate achieved at end of intervention
-
Functional exercise capacity: six‐minute walk distance or incremental shuttle walk distance
-
Endurance time at constant work: cycle or treadmill exercise, endurance shuttle walk
Secondary outcomes
-
Health‐related quality of life (HRQoL) measures : St George's Respiratory Questionnaire (SGRQ) or Chronic Respiratory Disease Questionnaire (CRDQ) or SF‐12 or SF‐36
-
Symptom scores
-
Muscle strength
Search methods for identification of studies
Electronic searches
Trials will be identified using the Cochrane Airways Group Specialised Register of trials, which is derived from systematic searches of bibliographic databases including the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, CINAHL, AMED, and PsycINFO, and handsearching of respiratory journals and meeting abstracts. All records in any language in the Specialised Register coded as 'COPD' will be searched using the following terms:
(exercis* or train* or "lower limb" or leg* or cycle or cycling OR bicycle OR walk* OR treadmill or *ergometer)
AND
(intens* or endur* or aerobic or anaerobic or interval* or intermittent* or continuous* or discontinuous*)
Searching other resources
Reference lists of all primary studies and review articles will be handsearched for qualifying studies. Authors of identified trials will be contacted and asked to identify other published and unpublished studies if necessary. Experts in the field will also be contacted.
Data collection and analysis
Selection of studies
Two reviewers will independently examine titles and abstracts of studies identified in the literature searches to remove obviously irrelevant studies and categorise the studies as follows:
1. INCLUDE: Study categorically meets all review criteria
2. EXCLUDE: Study clearly does not meet all review criteria
3. UNSURE: Study appears to meet some review criteria but insufficient information can be gleaned to categorically determine relevance
The two reviewers will independently retrieve and examine full text copies of studies in category 3 for compliance with the eligibility criteria. When crucial data that can determine the decision to include or exclude is missing, authors of the studies will be contacted to provide details of the missing data. Disagreements will be resolved by consensus and a third independent reviewer will be consulted when any disagreement is unresolved. A full record of decisions to include or exclude studies will be kept. Agreements between reviewers on study inclusion will be measured using Kappa statistics (ĸ value).
Data extraction and management
Two reviewers will independently use the full text copies of the included studies and extract data using standard data extraction forms. Data collected will include characteristics of the included studies (methods, participants, interventions and outcomes). If there are two or more detailed reports of the same study, data extraction will be performed separately for these articles and collated into a single data extraction form. When disagreement cannot be resolved by consensus, a third independent reviewer will be consulted. Where data is missing, authors of the studies will be contacted to provide details of the missing data. Finalised data will be entered into Review Manager by the primary reviewer (RZ), with random checks on accuracy.
Assessment of risk of bias in included studies
The following items will be assessed using the Cochrane Collaboration's 'Risk of bias assessment' tool Higgins 2008:
-
Sequence generation
-
Allocation concealment
-
Blinding of participants, investigators and outcome assessors
-
Incomplete outcome data
-
Selective outcome reporting
-
Other sources of bias
Measures of treatment effect
For continuous variables, the mean difference (MD) with 95% confidence interval (CI) or standardised mean difference (SMD), if different scales have been used, will be calculated in each study. Change‐from‐baseline scores will be preferred to final scores as change scores reduce inter‐subject variance. If possible, for each outcome, the common treatment effect and the limits of the CI will be compared to the minimal clinically important difference (MCID), where known, to determine the extent of the clinical benefit. The MCID is defined as "the smallest difference in score corresponding to the smallest difference perceived by the average patient that would mandate, in the absence of troublesome side effects and excessive cost, a change in patient management" Jaeschke 1989. The following MCIDs will be used:
-
10watts for peak work rate on a cycle ergometer Sutherland 2005
-
35 metres for six‐minute walk distance Puhan 2008
-
47.5 metres for incremental shuttle walk distance Singh 2008
-
100‐200 seconds for endurance cycle time Laviolette 2008
-
4 points improvement for the total score of SGRQ Jones 1992
-
0.5 point improvement in each item of CRDQ Jaeschke 1989
-
1 point improvement in Borg dyspnoea score Ries 2004.
If any of the outcome measures are dichotomous, the odds ratio (OR) with 95% CI will be calculated for each study.
Unit of analysis issues
For studies with multiple intervention groups, relevant intervention groups will be combined into a single group while relevant control groups will be combined into another single group so that a single pair‐wise comparison can be made with minimal unit‐of‐analysis error.
Dealing with missing data
Whenever possible, we will correspond with the original investigators to request missing data. When data are assumed to be missing at random, we will ignore the missing data and analyse the available data. Missing standard deviations of change scores can be obtained from other available information, such as standard error and confidence intervals, which can be converted to standard deviations. Sensitivity analyses will be performed to assess how sensitive the results are to changes or assumptions made with missing data.
Assessment of heterogeneity
Heterogeneity among studies will be assessed using the Chi2 test with an alpha of 0.05 used for statistical significance and with the I2 test (Higgins 2003).
Assessment of reporting biases
If necessary, funnel plots will be performed to assess reporting bias. Sensitivity analyses will also be undertaken to assess the influence possible bias has on the analysis.
Data synthesis
Separate meta‐analyses for studies that satisfy criteria (a), (b), (c) and (d) as described in Types of Intervention will be performed. Data in each categorical analysis will be pooled using random‐effects model as treatment effects between studies are expected to vary. The random effects model incorporates between study heterogeneity into the meta‐analysis.
Subgroup analysis and investigation of heterogeneity
Subgroup analysis will be performed to explore possible sources of heterogeneity when significant heterogeneity is present. Two subgroups analyses are identified a priori:
1. Disease severity: Treatment effects may vary according to the severity of the disease, i.e mild, moderate, severe and very severe, as defined by the GOLD guidelines (mild COPD characterised by FEV1 ≥ 80% predicted; moderate by 50% ≤ FEV1 < 80% predicted; severe by 30% ≤ FEV1 <50% predicted; and very severe FEV1 < 30% predicted) GOLD 2008.
2. Mode of exercise: Walking and cycling exercise training may induce different physiological and symptom responses.
