Content, uptake and adherence of exercise interventions after an acute exacerbation of COPD: a scoping review

Abstract

Introduction

Pulmonary rehabilitation is underutilised in patients after an acute exacerbation of COPD (AECOPD). Retrieving information regarding the setting, training modalities and the uptake and adherence to exercise interventions for these individuals in a vulnerable state could potentially guide future research.

Aim

To provide a comprehensive review of the existing literature on the content, uptake and adherence of different exercise interventions for patients after an AECOPD.

Methods

Eight different databases were searched for 1) patients experiencing an AECOPD and 2) performing any form of exercise intervention. Information on content, uptake and adherence was collected and the Consensus on Exercise Reporting Template (CERT) checklist was performed for each included record.

Results

59 distinct interventions were identified between 1998 and 2023 including a total of 9238 patients. All studies included patients requiring hospitalisation for the AECOPD, four studies additionally included patients not requiring hospitalisation for the AECOPD. Nine different settings were identified, with the majority of studies conducted in an inpatient setting (n=26) and including whole-body and strength exercises. The overall uptake was mentioned in 38 (62%) studies and was 70% with a 13% dropout rate. No paper reported the full CERT checklist. Adherence was defined a priori in 16 (27%) studies, with the most common definition being attendance of >80% of sessions.

Conclusion

Studies properly reporting on the uptake and adherence of well-described interventions, including information regarding fidelity, are needed to further investigate suitable programmes for patients experiencing an AECOPD.

Shareable abstract

Most exercise interventions following an AECOPD (59 identified) primarily focused on short inpatient programmes including whole-body and strength exercises. 49% of these interventions provided detailed content descriptions that would allow replication. https://bit.ly/4e54f6s

Introduction

An acute exacerbation of COPD (AECOPD) is defined as: “an event characterised by increased dyspnoea and/or cough and sputum that worsens in less than 14 days which may be accompanied by tachypnoea and/or tachycardia and is often associated with increased local and systemic inflammation caused by infection, pollution, or other insult to the airways” [1]. These AECOPDs characterise the disease and are part of its natural progression [1]. During an AECOPD, an acute worsening of the patient's muscle strength, exercise capacity, physical activity and symptoms is observed [2–6]. This systemic impact is observed after an emergency room visit or hospitalisation, but also in patients treated in a primary care setting [3, 5]. Management of AECOPDs includes medical therapy as well as management of their systemic consequences and interventions to prevent subsequent events [1]. Providing an exercise intervention as soon as possible, as part of a pulmonary rehabilitation programme, has been researched extensively.

Pulmonary rehabilitation after experiencing an AECOPD has emerged as a pivotal component of care and is widely recommended [7]. A Cochrane review showed that pulmonary rehabilitation is safe and effective and current American Thoracic Society (ATS) guidelines recommend it after an AECOPD [7–9]. The Cochrane review and the updated ATS guidelines included 17 randomised controlled trials and reported overall improvements in health-related quality of life, exercise tolerance, hospital readmissions and mortality for patients following pulmonary rehabilitation after an AECOPD [7, 8]. However, all the studies only included patients after hospital discharge and the programmes were heterogeneous in terms of setting and training modalities.

Despite its robust efficacy, pulmonary rehabilitation remains widely underused. Low numbers of referral, uptake and completion were observed for pulmonary rehabilitation programmes after an AECOPD. Referral ranges from 31 to 39%, with uptake ranging from 1.9 to 66% and completion from 10 to 73% [10–12]. These results originate from national audits and may differ from clinical studies. Additionally, these audits only included ambulatory or inpatient centre-based programmes. Alternatives such as telerehabilitation and home-based rehabilitation have already been explored in stable individuals with COPD to tackle the current challenges of programme uptake and adherence [13, 14]. An overview of rehabilitation options and their rates of uptake and adherence after an AECOPD could guide clinicians in selecting an appropriate programme for their patients. It could also be used as a benchmark for current clinical practice and inspire additional interventions. Various rehabilitation programmes for patients following an AECOPD have been studied over the past few years [7, 8, 15]. However, the heterogeneity and lack of detailed information on the interventions make it challenging to offer a suitable programme for patients in such a vulnerable state. A review on the reporting of nonpharmacological interventions in chronic diseases found that information across different domains of the intervention is often limited, which makes clinical applicability and the interpretation of results difficult [16]. The Consensus on Exercise Reporting Template (CERT) was developed to establish a consensus on the essential items that should reported in exercise interventions, thereby providing a framework for researchers to report the intervention [17]. This tool could useful for evaluating how extensively an intervention is described in clinical studies.

This scoping review aims to systematically map and comprehensively review the existing literature on the content, uptake and adherence of different training modalities for patients after experiencing an AECOPD. Additionally, we aim to report the interventions according to the CERT items to assess applicability. This may help in designing future interventions in the context of AECOPDs for different patient types and healthcare settings.

Methods

This scoping review was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis extension for scoping reviews (PRISMA-ScR) and Covidence was used for deduplication, screening and data extraction [18]. Citation searching was reported according to the Terminology, Application, and Reporting of Citation Searching (TARCiS) statement [19].

Search methods and eligibility criteria

A literature search was conducted with the assistance of a biomedical information specialist (T.V.). Eight electronic databases were searched for relevant literature, namely CINAHL (via EBSCOhost – coverage from 1981 to search date), the Cochrane Database of Systematic Reviews and Cochrane Central Register of Controlled Trials (via Cochrane Library – from inception (date unknown) to search date), Embase (via www.embase.com – from 1974 to search date), PEDro (Physiotherapy Evidence Database) (via pedro.org.au – from 1929 to search date), PubMed (via the National Center for Biotechnology Information, including Medline − from 1946 to search date), SPORTDiscus (via EBSCOHost – from 1800 to search date) and Web of Science (via webofscience.com; SCI-EXPANDED – from 1955 to search date, SSCI – from 1956 to search date, AHCI – from 1975 to search date, CPCI-S – from 1990 to search date, CPCI-SSH – from 1990 to search date, BKCI-S – from 2005 to search date, BKCI-SSH – from 2005 to search date, ESCI – 2018 to search date). There was no year or language barrier set in the search strategy. Any non-English full-text papers were translated in Deepl Translator (DeepL GmbH, Cologne, Germany). A first literature search was performed on 27 October 2021 and an update was performed on 7 June 2022, prior to initiating the full-text screening. The search strategy was based on the following two concepts: 1) the presence of an AECOPD and 2) any form of exercise training. Additional papers deemed relevant by the researchers were manually added based on forward citation searching. All identified papers were exported to Covidence for screening and data extraction. An overview of the search strategy for each database can be found in supplement S1.

Studies reporting an exercise intervention for patients after experiencing an AECOPD were considered for eligibility. An AECOPD was defined based on the author's description. Additionally, no specific time window between the AECOPD and the initiation of the intervention was required; this was also based on the description of the authors. Records were excluded from full-text screening based on the following criteria: 1) review papers, protocol papers, guidelines, letters to editors, congress reports or abstracts; 2) interventions lasting on average less than 7 days, given the assumption that many studies would be performed only during a hospital stay; 3) if the intervention consisted only of respiratory chest physiotherapy or breathing exercises; and 4) when only general advice or education on exercises or physical activity was given.

After removing duplicates in Covidence, two independent researchers (M.W. and J.D. or J.D.R.) screened the records by title and abstract based on the eligibility criteria. Both researchers were blinded to each other's responses. Disagreements between the two researchers were discussed in groups with senior researchers (T.T. and H.D.) to make a final decision in consensus. The full-text screening was performed in the same manner.

Data extraction

A data-extraction template was developed in Covidence by M.W., T.T. and H.D. and can be found in supplement S2. Data from the included records were then extracted by two independent researchers (M.W. or Y.C. and J.D. or J.D.R. or P.M.). Covidence revealed disagreements after the data extraction, which were openly discussed with senior researchers (T.T. and H.D.) to make a final decision.

The following information was extracted from every included record: 1) general descriptive information regarding the study; 2) information on the number of patients eligible, included and dropping out of the intervention; 3) adherence to the intervention; and 4) content of the exercise intervention based on the CERT checklist. General information included the first author's name, title, year of publication, study design, primary outcome and effectiveness, and baseline characteristics of the study sample. Patients were classified as having a moderate or severe AECOPD based on the criteria of the 2022 Global Initiative for Chronic Obstructive Lung Disease report [20]. Eligibility was defined as the number of patients suitable for participation in the study, uptake was defined as the percentage of patients taking up the offer of starting the study and dropout was defined as the number of patients defined as dropout in the study protocol. To assess the adherence to the intervention, we collected information on its explicit mention or definition in the study protocol. Lastly, the reporting of the content of each intervention was evaluated using the 16-item CERT checklist, which entails the following components: 1) the material used; 2) the provider of the programme; 3) nine items regarding the delivery of the intervention; 4) the location of the intervention; 5) the description of the exercises of the intervention; 6) two items on intervention tailoring; and 7) one item on whether the intervention was delivered and performed as intended [17]. The mention of each item in every included record was collected. The delivery of the intervention included whether exercises were performed individually or in groups, whether exercises were supervised, details on motivation strategies, how the intervention progressed, detailed information on the exercises, the content of any home programme, nonexercise components and how adverse events were documented and managed. Any mention of these items in the studies was collected to provide an overview of clinical applicability. Additionally, specific information was collected regarding the setting of the intervention, materials used, the focus of the exercises, nonexercise components, frequency, intensity, duration of the sessions and intervention, possible maintenance follow-up programmes, and intervention adherence. The focus of the intervention was grouped according to whole-body exercises (including the exercise of large muscle groups that substantially increase whole-body metabolism and ventilation such as cycling, walking or step exercises), strength exercises or other exercises (e.g. inspiratory muscle training, respiratory physiotherapy and electro-muscle stimulation).

Data analysis

The data was charted for both the study characteristics and the exercise intervention. The samples from the studies were pooled together to provide an overview of screening, uptake and completion across different intervention settings. Weighed means were calculated. An overall overview of the different components of the exercise intervention was established and the proportion of mentioned items related to the content of the intervention in the included records was collected [17]. Considering that the guidelines for exercise reporting were published in 2016, we only performed a sub-analysis on papers published after 2016 to observe differences in reporting. No meta-analysis was performed as the data were descriptive and a high level of heterogeneity was assumed.

Results

Description of included studies

A total of 13 669 papers were identified from the eight databases across the two searches. After deduplicating, 7826 records were screened for title and abstract, of which 142 were screened for full text. Figure 1 presents the reasons for the exclusion of full-text articles. 62 records were included in this review with 59 unique interventions. Two studies published short and long-term effects in two separate papers of the same patient population and intervention; so for the purpose of this review, the data was merged as one study [21, 22]. One non-English paper was included and translated from Russian [23].

FIGURE 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) 2020 flow diagram. CINAHL: Cumulative Index to Nursing and Allied Health Literature; PEDro: Physiotherapy Evidence Database.

Studies were conducted in 24 different countries of which the majority were carried out in Spain (n=10, 16% of studies) and the United Kingdom (n=9, 15% of studies). The studies were conducted between 1998 and 2023 of which half were published between 2015 and 2023. One paper from 2022 and one from 2023 were manually added [24, 25]. 23 studies (37%) were published in the last 5 years. The majority of studies were randomised controlled trials (n=41, 67%). Other study designs included retrospective studies (n=5), prospective intervention studies (n=7), quasi-randomised controlled trials (n=2), two prospective cohort studies, two feasibility studies, one case study and one pilot study. A total number of 9238 patients were included across 117 groups. All studies included patients after a severe AECOPD, four studies additionally included patients after a moderate AECOPD alongside those with a severe AECOPD. Individuals with a moderate AECOPD represented 76.0, 72.5 and 73.9% of the sample [26–28]. One study did not report on the proportion of patients included with a moderate AECOPD [29]. The mean duration of the study was 14 (range 1–52) weeks. A weighted mean±sd age of 70±3 years was observed at baseline, with 69% male participants with a weighed mean forced expiratory volume in 1 s of 44±8% pred. Table 1 includes the primary outcome and the effectiveness of the intervention for each included study.

TABLE 1.

Overview of different components, study design with the mentioned Consensus on Exercise Reporting Template (CERT) items and the outcomes with the effectiveness of the rehabilitation programme included in each record

Study, year	Duration (weeks)	Included (n)	Focus of exercise programme			Supervision	Information on intensity	Description of intensity	Information on progression	Nonexercise component	Design of the study (CERT (n))	Primary outcome
			WBE	STE	Other
Inpatient rehabilitation
Abdellaoui [35], 2011	6	17			✓	NR	✓	Tolerance	✓	✓	RCT (11)	Quadriceps force# and muscle oxidative stress
Borges [46], 2014	1	46		✓	✓	✓	✓	% RM	✓		RCT (14)	Exercise capacity#, HRQoL, lower limb muscle force# and adverse events
Chaplin [50], 2013	1	29	✓		✓	✓	✓	Symptoms	✓		RCT (12)	Quadriceps force¶
Cilione [73], 2002	2	132	✓	✓		✓		NR		✓	Prospective intervention study (7)	6MWD
Clini [29], 2009	4	1826	✓	✓		✓	✓	% maximum load	✓	✓	Retrospective study (11)	Feasibility, 6MWD#, dyspnoea and HRQoL
Garuti [74], 2003	2	149	✓	✓		✓		NR		✓	Prospective intervention study (9)	Anxiety and depression¶
Greulich [75], 2014	1.3	49	✓		✓	✓	✓	Frequency and amplitude	✓		RCT (9)	Safety, exercise capacity# and QoL#
He [33], 2015	1.3	101	✓	✓	✓	✓	✓	Symptoms	✓	✓	RCT (11)	Safety, 6MWD¶, CAT¶, symptoms¶ and activities of daily life¶
Kirsten [34], 1998	1.5	29	✓			✓	✓	Walking distance	✓		RCT (10)	6MWD¶, lung function¶ and blood gas
Knaut [76], 2020	1	26	✓			NR		NR	✓		RCT (6)	Inflammatory markers
Kofod [47], 2017	1	34		✓		✓	✓	% RM	✓		Prospective intervention study (13)	Training load¶ and quadriceps force¶
López-López [77], 2018	1.5	39	✓	✓	✓	✓	✓	Tolerance			RCT (6)	5×STS# and symptoms#
López-López [78], 2020	1.5	66	✓		✓	✓	✓	Heart rate and SpO2		✓	RCT (8)	Functionality# and HRQoL#
López-López [32], 2021	1	48		✓	✓	✓		NR			RCT (8)	Quadriceps force#, symptoms# and adverse events
López-López [79], 2021	1.5	43	✓		✓	✓	✓	Borg score	✓		RCT (11)	Quadriceps force#, balance#, HRQoL#, adverse events and adherence
Meglic [54], 2011	1	19		✓	✓	✓		NR			Prospective intervention study (5)	Symptoms¶, HRQoL¶ and feasibility (qualitative assessment)
Mirza [49], 2020	1	38	✓	✓	✓	✓/X	✓	% 2MWT, repetition number	✓	✓	RCT (13)	2MWD#, quadriceps force#, 30 s STS, TUG and physical activity
Shimoda [80], 2021	2	24	✓	✓		✓	✓	% ISWT		✓	Prospective intervention study (5)	ISWT¶
Torres-Sánchez [81], 2016	1	49	✓	✓	✓	✓	✓	Symptoms			RCT (7)	Lung function, quadriceps force#, handgrip strength, 2MSP#, dyspnoea, QoL# and HADS#
Torres-Sánchez [82], 2017	2	58	✓			✓	✓	Borg score	✓		RCT (10)	Quadriceps force#, balance# and 30 s STS
Torres-Sánchez [83], 2018	2	90		✓		✓	✓	Symptoms	✓		RCT (8)	Health status#
Troosters [52], 2010	1	40		✓	✓	✓	✓	% RM	✓		RCT (13)	Quadriceps force#
Van Dam Van Isselt [36], 2014 Van Dam Van Isselt [37], 2019	6	1) 61 2) 58	✓	✓	✓	NR		NR		✓	1) Prospective intervention study (5) 2) Prospective cohort study (5)	1) Functional capacity¶ and health status¶ 2) Disease-specific health status#
Vitacca [27], 2021 Vitacca [26], 2021	4	1) 1057 2) 2066	✓			✓	✓	% 6MWD	✓	✓	1) Retrospective study (9) 2) Retrospective study (8)	1) 6MWD¶ 2) Respiratory ICF Maugeri core set¶
Vivodtzev [56], 2006	4	17	✓		✓	✓	✓	Tolerance	✓	✓	RCT (14)	Quadriceps force#, 6MWD and QoL
Yohannes [63], 2003	1	110	✓			✓	✓	Tolerance			RCT (11)	Barthel index and Borg scale
Outpatient rehabilitation
Chaplin [84], 2015	7	75	✓	✓		✓/X	✓	% RM, Borg score	✓	✓	Retrospective study (9)	CAT
Deepak [59], 2014	12	60	✓	✓	✓	✓	✓	Baseline function		✓	RCT (8)	6MWD# and HRQoL#
Ko [61], 2011	8	60	✓	✓	✓	✓	✓	Tolerance	✓	✓	RCT (11)	Healthcare resource utilisation and health status#
Man [48], 2004	8	42	✓	✓		✓/X		NR		✓	RCT (7)	ISWT#, health status# and HRQoL#
Riario-Sforza [85], 2005	4	37	✓	✓	✓	✓	✓	Patient's capacity		✓	Prospective cohort study (4)	6MWD and lung function
Seymour [62], 2010	8	60	✓	✓		✓	✓	Baseline function		✓	RCT (6)	Hospital admission# and ER visits#
Zaripova [23], 2020	40	140	✓		✓	✓	✓	% heart rate	✓	✓	Quasi RCT (9)	Symptoms¶ and physical performance¶
Home-based rehabilitation
Ghanem [86], 2010	8	39	✓	✓	✓	✓/X		NR		✓	RCT (6)	Exercise tolerance# and HRQoL#
Johnson-Warrington [87], 2016	12	78	✓	✓		✓/X	✓	Low-intensity exercises	✓	✓	RCT (12)	Hospital readmissions
Lalmolda [57], 2017	8	50	✓	✓	✓	✓		NR	✓	✓	Case–cohort study (7)	Healthcare utilisation#
Murphy [88], 2005	6	31	✓	✓		✓/X	✓	Borg score	✓		RCT (9)	Exercise capacity¶, QoL¶, muscle strength and exacerbations
Wageck [60], 2020	8	15	✓	✓		✓/X	✓	Borg score, % 6MWT		✓	Feasibility study (11)	Barriers to uptake and completion
Inpatient and home-based rehabilitation
Guell-Rous [51], 2020	12	53	✓	✓	✓	✓/X	✓	Targeted repetition number	✓	✓	RCT (14)	Exacerbations, symptoms#, 6MWD, HRQoL, functional status# and adverse events#
Behnke [21], 2000 Behnke [22], 2003	24	1) 46 2) 26	✓		✓	✓/X	✓	Treadmill walking distance	✓		RCT (11)	1) Lung function, exercise performance# and symptoms# 2) Hospital admissions# and bronchodilator use#
Lu [53], 2020	10	82		✓		✓/X	✓	Targeted repetition number			RCT (9)	CAT#, 6MWD# and symptoms#
Öncü [55], 2017	3	82			✓	✓/X	✓	Tolerance		✓	RCT (8)	Lung function, 6MWD# and QoL
Orme [30], 2018	2	33			✓	X		NR		✓	Feasibility RCT (12)	Feasibility and acceptability
Inpatient and outpatient rehabilitation
Ali [89], 2014	3	30	✓	✓	✓	NR	✓	% RM, % CPET	✓	✓	Quasi RCT (8)	Symptoms, 6MWD#, VʹO2max# and QoL#
Eaton [43], 2009	8.4	97	✓	✓		✓		NR		✓	RCT (7)	Health care utilisation
Tung [90], 2020	12	44	✓	✓	✓	NR		NR		✓	RCT (3)	Lung function, 6MWD#, symptoms, QoL and hospital admissions
Outpatient and home-based rehabilitation
Ko [44], 2021	8	136	✓	✓	✓	✓/X	✓	Heart rate	✓	✓	RCT (11)	Hospital readmissions# and physical activity
Revitt [58], 2013	7	160	✓	✓	✓	✓/X	✓	Peak VʹO2, Borg score	✓	✓	Prospective intervention study (10)	ISWT¶, ESWT¶, health status¶ and hospital readmissions¶
Home-based rehabilitation
Hornikx [91], 2015	4	30			✓	X	✓	Step goal	✓	✓	RCT (12)	Physical activity
Valeiro [25], 2022	12	46			✓	X	✓	Step goal	✓	✓	RCT (13)	Physical activity# and sedentary behaviour
Community care
Machado [31], 2020	3	23	✓	✓	✓	✓	✓	Heart rate, Borg score	✓	✓	Quasi-experimental pilot study (13)	Symptoms, quadriceps force#, 5×STS and health status#
Tele-rehabilitation
Polo [24], 2023	8	209	✓		✓	✓	✓	Exercise capacity	✓	✓	RCT (12)	Hospital readmissions and mortality
Outpatient or inpatient rehabilitation
Carr [28], 2009	3	34	✓	✓	✓		✓	Symptoms		✓	RCT (7)	HRQoL and 6MWD
Puhan [41], 2012	12	36	✓	✓		✓		NR		✓	RCT (4)	Exacerbation rate
Outpatient or home-based rehabilitation
Ko [38], 2014	16	185	NR	NR	NR	✓/X		NR		✓	Retrospective study (7)	Hospital readmissions¶
Ko [39], 2017	8	180	✓	✓	✓	✓/X	✓	Tolerability and vital signs	✓	✓	RCT (13)	Hospital readmissions#
Kjaergaard [40], 2020	7	150	✓	✓		✓	✓	% RM, Borg score		✓	RCT (13)	Mortality and hospital readmissions
Home rehabilitation versus inpatient rehabilitation
Cox [45], 2018	3	58	✓			✓	✓	% RM	✓	✓	RCT (15)	Feasibility, 6MWD and costs#
Setting not reported
Daabis [42], 2017	8	45	✓	✓			✓	% RM, % 6MWT	✓	✓	RCT (9)	Symptoms, lung function, 1 RM#, 6MWD and HRQoL

^#: Significant (p<0.05) between-group difference. ^¶: Significant (p<0.05) within-group difference. 2MSP: 2-min step-in-place; 2MWD: 2-min walking distance; 2MWT: 2-min walking test; 6MWD: 6-min walking distance; 6MWT: 6-min walking test; CAT: COPD assessment test; CPET: cardiopulmonary exercise testing; ER: emergency room; ESWT: endurance shuttle walk test; HADS: hospital anxiety and depression score; HRQoL: health-related quality of life; ICF: International Classification of Functioning, Disability and Health; ISWT: incremental shuttle walk test; NR: not reported; QoL: quality of life; RCT: randomised controlled trial; RM: repetition maximum; S_pO2: oxygen saturation measured by pulse oximetry; STE: strength exercise; STS: sit-to-stand; TUG: timed up-and-go test; V′_O2: oxygen uptake; V′_O2max: maximal oxygen uptake; WBE: whole-body exercise.

Information on uptake and completion

38 studies (62%) reported the number of patients screened for eligibility in the study, of which 35 (57% of total) studies reported the number of patients that were eligible for inclusion. The overall flow of the number of patients screened, the number of patients eligible and the total number of patients included is shown in figure 2. An overall uptake of 70% was observed in the 35 studies. The study by Orme et al. [30] reported the lowest observed uptake of 29%. An uptake of 100% was reported in two studies [31, 32]. 59 studies (97%) reported the number of patients at baseline and after the intervention. Two studies only reported the number of participants at baseline and the total number of dropouts in both groups together and were therefore not incorporated in this analysis [33, 34]. The overall completion rate of all participants was 91%, with 93% in the intervention group and 87% in the control group. The proportion of patients dropping out of the study ranged from 0 to 46% with a mean of 13±13% in the control groups and 0 to 56% with a mean of 13±13% in the intervention groups. For interventions in an inpatient setting (n=26), a mean dropout rate was observed of 13% (range 0–44%) and 9% (range 0–31%) in the control and intervention groups, respectively. For an outpatient setting (n=6) this was 8% (range 0–13%) and 14% (range 0–23%), for a home-based setting (n=5) this was 5% (range 0–13%) and 11% (range 0–18%) and for other settings (n=21) this was 15% (range 0–47%) and 19% (range 0–56%). The largest proportion of dropouts in an intervention group was 56%, in a telerehabilitation programme [24].

FIGURE 2.

Flow of screening, eligibility, uptake and completion of included papers. N: number of studies; n: number of patients.

Components of the exercise intervention

Various settings of providing rehabilitation for patients after an AECOPD were observed, of which the majority was an inpatient setting (n=26, 46%) with a typical duration of ≤4 weeks. Only two inpatient interventions reported a duration of 6 weeks [35–37]. Other reported settings were outpatient (n=7), home-based (n=5), home-based following inpatient rehabilitation (n=5), outpatient following inpatient rehabilitation (n=3), home-based following outpatient rehabilitation (n=2), at home (n=2), community care (n=1) and telerehabilitation (n=1). Three studies reported patients who followed rehabilitation either on an outpatient or home-based basis [38–40]. In two studies, patients could choose between outpatient or inpatient rehabilitation [28, 41]. One study did not mention the setting of the intervention [42]. The average duration of the intervention was 5 weeks (range 1–40 weeks). Six studies provided a maintenance follow-up intervention after the initial intervention [22, 38, 39, 43–45]. Table 1 shows the focus of the exercise programme, information on the supervision, intensity, progression, if there were nonexercise components, if there was a maintenance follow-up programme, and the primary outcome of each study.

The different interventions showed large heterogeneity in the rehabilitation components. Figure 3 shows an overview of all components that were reported in terms of the focus, materials, intensity, progression, frequency and nonexercise components across all interventions. More detailed information for each study per component is provided in table S3.

FIGURE 3.

Overview of different components of rehabilitation included in the review. The number in the bold line identifies in how many studies this was mentioned. The number per smaller item states the frequency of reporting each item for all mentioned studies. 6MWD: 6-min walking distance; ISWT: incremental shuttle walk test; NMES: neuro-muscular electrostimulation; TENS: transcutaneous electrical nerve stimulation; Vʹ_O2peak: peak oxygen consumption; RM: repetition maximum. More detailed information is provided in table S3.

The majority of the interventions included both whole-body exercises and strength exercises (n=31, 52%). 13 studies focused only on whole-body exercises and two interventions only on strength exercises. Other additional focuses could be respiratory muscle training, neuromuscular electrostimulation, whole-body vibration, physical activity, sedentary behaviour and transcutaneous electrical nerve stimulation.

Adherence

The adherence to the intervention was neither mentioned nor defined in 30 (49%) studies. 14 studies (23%) mentioned adherence to the intervention but did not use an a priori definition. This was reported as 1) the percentage of sessions patients followed, ranging from 92 to 97% [35, 46–49], 2) the total number of sessions completed [39, 44, 50–52] or 3) the percentage of individuals who completed the intervention [53–56]. 16 (27%) studies defined adherence and mentioned it in the paper. Nine distinct definitions delineated the proportion of attendance to the intervention or the attendance at the subsequent follow-up measurement. The two most frequently identified definitions were an attendance >80% to the sessions (n=3, 19%) [38, 45, 57] and an attendance at the follow-up (n=2, 19%) [58, 59]. Adherence was defined as exceeding 70% [60, 61] or 75% [40, 43] of the attendance rates. Other definitions were based on the attendance thresholds, qualitative interviews with the assessors, including more than one session, >50%, >60%, >85% or 100% attendance [24, 27, 29, 31, 42, 62, 63]. Each of these definitions was reported once. Definitions were not related to the respective duration of the intervention.

Applicability of the CERT checklist

Figure 4 shows the reported intervention details as a percentage of all studies. A mean of nine out of 16 items (range 4–15) was reported in the studies. The replication of exercises of the intervention (item 8) would not be possible in 51% of the individual studies and specific information on sets, repetitions, duration and intensity (item 13) was lacking in 61%. Only 11 studies (18%) reported information regarding the fidelity of the programme, i.e. if the exercise was delivered and performed as planned. No differences were observed in the reporting of items in records published after 2016 (figure S4).

FIGURE 4.

Overview of the different items of the Consensus on Exercise Reporting Template (CERT) checklist and how often they are mentioned in the included records as a percentage of all records.

Discussion

This scoping review, identifying 62 records and 59 different interventions, provides an overview of content, uptake and adherence of the available literature on exercise interventions for people experiencing an AECOPD. The majority of these studies were executed in an inpatient facility and mainly focused on whole-body and strength exercises. 57% of the studies reported on uptake and most studies included patients requiring hospitalisation for the AECOPD. The content of each intervention was heterogeneous and often lacked sufficient information to ensure reproducibility and verify the effectiveness of the results.

Pooled data of the number of participants screened and included and who completed the exercise intervention showed an overall uptake of 70.1% (35 studies), an overall completion rate of 92% (59 studies) and a completion rate of 93% in the intervention group. This was higher than data reported in national audit studies where uptake in outpatient pulmonary rehabilitation ranged from 1.9 to 66% and completion from 10 to 73% [10–12]. These audit studies reported the uptake and completion in an inpatient or outpatient pulmonary rehabilitation facility. The studies that were identified in this scoping review generally reported on shorter exercise interventions in an inpatient hospital setting when patients were hospitalised for their AECOPD or other settings. Additionally, these studies were not necessarily a pulmonary rehabilitation programme, but an exercise intervention, making this comparison with audit data difficult. There was only one pulmonary rehabilitation study in this review executed in an outpatient setting describing numbers of uptake (53%) and completion (85%), more in line with the audit studies [61]. The largest proportion of dropouts was observed in a telerehabilitation setting, targeting patients with health disparities and only included African American and Hispanic patients [24]. We found no difference in completion rates among other settings, although this comparison is biased by the large heterogeneity in the intervention components. Future audit studies reporting the uptake and adherence of pulmonary rehabilitation in various real-life settings after an AECOPD could offer greater insights into patient preferences and willingness across different settings after an AECOPD.

To tackle these low numbers of uptake and adherence, innovative and sustainable pulmonary rehabilitation interventions may be needed [64]. The present results show that nine different settings or combinations of settings were trialled. Exercise interventions performed in an inpatient setting were the most commonly investigated. Often these interventions were initiated during hospital admission for an AECOPD, which allows for easier recruitment. However, no conclusions can be made yet as to the most beneficial setting since some settings were only investigated in a single study, such as community care or telerehabilitation [24, 31]. Additionally, rehabilitation in primary care has not yet been investigated but potentially holds promise given its high accessibility and flexibility [65]. One of the main reasons patients are not willing to follow outpatient rehabilitation after an AECOPD are transport issues, disruption of the daily routine and inconvenient timing [66]. Therefore, investigating innovative approaches to pulmonary rehabilitation across lines of healthcare in collaborative models seems important. Additionally, only six studies provided any form of maintenance follow-up intervention [22, 38, 39, 43–45]. Maintenance rehabilitation programmes have the potential to sustain training effects and may even lead to further improvements in terms of clinical outcomes [67]. This merits further study. Lastly, the identified studies are nested in an existing healthcare system. Hence healthcare system research around rehabilitation after AECOPD may help to understand the barriers and opportunities of embedding interventions in a specific healthcare system [68].

Only four studies also included patients experiencing a moderate AECOPD [26–29], of which two studies performed sub-analyses [26, 27]. These two studies showed a larger effect of the intervention on functional exercise capacity in patients with a severe AECOPD compared to patients with a moderate AECOPD [26, 27]. Nevertheless, it seems important to be inclusive in this patient population for future research since the impact of an AECOPD on exercise capacity, muscle strength and physical activity is comparable after moderate or severe AECOPDs [3, 5].

The 59 identified interventions show a variety of programme components and outcome measures. This makes systematic reviews on this topic challenging and caution should be taken when performing meta-analyses. The majority of the identified studies lacked information on motivation strategies (83%), exercise intensity (66%), fidelity (76%) and whether the exercises were performed individually or in groups (69%). No differences were observed in reporting the intervention content between papers published before or after the release of the CERT guidelines in 2016 [17]. The absence of fidelity reporting can threaten the internal validity of the studies. This makes it challenging to interpret the effectiveness of the intervention and its clinical applicability. Complete reporting of the exercise intervention using a standardised guideline in future studies or adding this as a requirement for journals could possibly mitigate these challenges. A minimum requirement to replicate the exercise interventions should be advised, as this was currently only possible in 49% of the studies (item 8). A CORE-CERT checklist of seven items was developed based on the CERT items as a minimum requirement for reproducibility in exercise studies for breast cancer patients in clinical practice [69]. This list could also be used for patients with chronic respiratory disorders, including COPD, which is less extensive than the full CERT checklist and focuses on some of the main components that should be reported. Similarly, for outcomes, future clinical trials could consider including a core outcome set for pulmonary rehabilitation as a minimal requirement, which has been developed recently [70]. This core outcome set covers the relevant dimensions to be covered by the outcome measures, but is not yet very specific in terms of the exact outcome measures. In addition, the core outcome set does not include the measurement of AECOPDs, which might be important to add in this specific patient population.

17 identified studies (28% of the total studies) reported an a priori definition and mentioned adherence to the intervention. An extension of the CONSORT statement of nonpharmacological interventions requires that adherence is reported in studies [71]. Unfortunately, there is no consensus on the definition of adherence to exercise interventions. Two different definitions were used in this scoping review; more specifically, the number of sessions patients followed or the attendance at their follow-up assessment visit. The latter, however, does not give any information regarding how well the intervention was executed. Other definitions of adherence such as the duration of the intervention or the intensity were not reported but could be further studied [72]. Additionally, the concluded definition of adherence should be based on the outcome that is measured. Currently, the most frequently used definition of adherence was a minimum attendance of 80% to exercise sessions. More studies are required to identify the optimal cut-off or whether this should be seen as a continuous, dose–response type outcome.

Clinical implications

A large variety of exercise interventions have been investigated in individuals recovering from an AECOPD. This scoping review can provide some inspiration for clinicians looking to integrate a new intervention into clinical practice, although effectiveness should be considered. To that end, the present review could be combined with results of the systematic reviews on effectiveness of exercise interventions after an AECOPD. Importantly, the interventions could be reproduced by clinicians in fewer than 50% of the studies. This highlights the need for enhanced transparency in reporting exercise interventions across research studies. Additionally, when incorporating a new intervention, it is advisable to also assess adherence and fidelity, considering their importance in understanding the results of an intervention. Lastly, studies for people experiencing a moderate AECOPD are lacking and no guidelines currently exist for this specific population. This requires further attention.

Strengths and limitations

To our knowledge, this is the first scoping review to explore in depth the content and adherence of different exercise interventions for people following an AECOPD. It is the first scoping review to include all possible exercise interventions for people experiencing an AECOPD and to provide further information on uptake, adherence and the applicability of the CERT guidelines. We deliver an extensive summarised overview of the available literature, allowing the opportunity for new systematic reviews. We extensively sought available literature from eight different databases and reported according to the PRISMA-ScR guidelines [18]. Additionally, non-English literature was translated and could therefore be included in the current review.

Despite our extensive search and assuring that each record was screened by two individuals, there is still a possibility that relevant studies could have been missed. No meta-analysis on the evidence of the acquired data was performed, given the high heterogeneity. Meta-analyses have confirmed that forms of rehabilitation have the potential to improve exercise tolerance, health-related quality of life and, notably, readmission rates. These systematic reviews, however, lumped programmes together and did not provide a similar level of detail on the content, uptake and adherence. This lack of detail limits their ability to guide further developments in alternative and accessible approaches to rehabilitation after exacerbations.

Conclusion

This scoping review identified 59 different interventions for patients experiencing an AECOPD. Rehabilitation programmes after acute exacerbations of COPD can take many shapes and forms, but high-quality studies with proper reporting on uptake and adherence of well-described interventions using standardised guidelines are needed to further investigate suitable programmes for patients experiencing an AECOPD.

Points for clinical practice and questions for the future

• A wide variety of exercise interventions with different components exist for patients following an AECOPD.

• The majority of interventions were performed in an inpatient setting and consisted of whole-body and strength exercises.

• Settings such as telerehabilitation or community care rehabilitation have only been investigated once and need further exploration.

• Future studies should also include individuals after a moderate AECOPD.

• Transparent and detailed reporting of the exercise intervention is advised and could be recommended by journals.

• Reporting the uptake and adherence of exercise interventions in different settings after an AECOPD can provide more insights into preference and willingness.

Supplementary material

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Supplementary material ERR-0172-2024.SUPPLEMENT ^{(711.3KB, pdf)}