One researcher (AM) performed an electronic literature search on PubMed, Embase and Web of Science from inception until January 2021. The following search strategy was used: ((COPD [title/abstract] OR chronic obstructive pulmonary disease [title/abstract/MeSH]) AND (hospital*[title] OR exacerbation [title/abstract])). The search results were imported to EndNote X9 (Clarivate Analytics, Philadelphia, PA, United States of America) and the duplicates were identified and removed.

A single researcher (AM or MSBG or CB) performed the title screening conservatively, i.e., excluding studies which clearly did not fulfil the criteria. Abstract and consequent full-text screening were performed independently by two out of three researchers (AM and CB or MSBG and CB). A consensus-based decision was made after discussion when discrepancies were present.

Studies were included if they (i) studied patients with COPD that suffered from at least one AECOPD throughout the study; (ii) were prospective cohort studies; (iii) performed at least one type of health status assessment; and (iv) were written in English. Patients had to be assessed at baseline (in a stable phase of the disease) and after a follow-up period. The follow-up assessment(s) could have been conducted immediately after an AECOPD, during the course of a single AECOPD, or over a more extended period of time with the onset of AECOPD during the follow-up period being recorded (e.g., to compare changes in health status in frequent exacerbators vs. non-frequent exacerbators). Given the variability of used definitions of AECOPD in the literature, for the purpose of this review, AECOPD could be defined using symptom-based (i.e., patient-reported worsening of respiratory symptoms either to a healthcare professional or using a diary or tool) or event-based (i.e., change in treatment – medication and/or hospitalization) definitions, or a combination of both.23

Studies reporting on the short-, mid- or long-term effects of any intervention were excluded, unless it was possible to retrieve the data of a control group receiving only standard of care. Abstracts in conference proceedings were also excluded.

Based on the sub-classification of health status previously proposed for patients with COPD,18 we included measures of physiological functioning (exercise tolerance, muscle function and body composition), complaints (subjective complaints, expected dyspnoea and dyspnoea emotions), functional impairment (subjective impairment, behavioural impairment and actual physical activity) and QoL (general QoL, health-related QoL, satisfaction and relations). Measures of pulmonary function were not included.

A customized data collection tool in Microsoft® Excel (Microsoft, Redmond, Washington, United States of America) and a data extraction table in Microsoft® Word (Microsoft, Redmond, Washington, United States of America) were developed to extract the most relevant information from the included studies and facilitate their subsequent analysis and interpretation. Data extraction was performed by AM and MSBG. Information on study design and timing of assessment, sample size, baseline characteristics (age, gender, body mass index (BMI), forced expiratory volume in one second (FEV1)), definition of (frequent) exacerbations used, AECOPD management and setting, measures of health status and main results regarding health status was collected.

Data are presented in a tabularized format with a narrative summary linking the review results with the aim and review question. Key findings were categorized according to the sub-domains of health status proposed by Vercoulen and colleagues.18 In addition, whenever possible, results synthesis were grouped on acute vs. long-term effects of AECOPD, and on single vs. repeated AECOPD. For the purposes of this review, acute effects combined the results found regarding the onset and first days of an AECOPD, long-term effects combined the data on recovery/sustained changes over time, i.e., data from post-AECOPD periods, annual changes and follow-up times. Numerical summaries for the definition of (frequent) exacerbations, AECOPD characteristics and management, and measures of health status were collated. Schematic overviews were further developed to provide a visual synthesis of the results.

The literature search provided a total of 23,677 records. After the removal of duplicates, 12,662 records were screened for relevant content using titles and abstracts. From these, 12,199 were excluded through title screening and 388 through abstract screening. Thus, the full text of 74 potentially relevant reports was assessed. Twenty-two studies were included in the review. The screening process is visualized in a flow chart (Fig. 1).

Figure 1.

PRISMA flowchart of the included studies.

(0.51MB).

Characteristics of the included studies are shown in Table 1. Studies were conducted between 2004 and 2020.

Most studies (n = 14) defined AECOPD based on an increase in respiratory symptoms24-37 leading to changes in medication (n = 12) (e.g., treatment with antibiotics or systemic steroids)26,28,30,31,33-35,37-41 or to hospitalization (n = 4).35,38,39,41 Follow-up time varied from six weeks up to eight years. Ninety-one percent (n = 20) of the studies25,26,28-45 involved a sample combining patients that suffered a single exacerbation and patients that suffered repeated exacerbations. Severity of exacerbations was usually not reported. Nevertheless, 80% of the studies reporting this information were focused on moderate to severe AECOPD.24,26,28,29,37,40,41,45 Most reported treatment was composed by antibiotics, oral steroids or a combination of both (n = 10).25,26,28,29,34,37-41 Twelve studies included a sample combining patients treated at home with patients treated in the hospital.26-29,33,36-42

Physiological functioning was assessed in five studies with the six-minute walking distance (6MWD),24,26,28,29 the quadriceps maximum voluntary contraction (QMVC)24,30 and the fat-free mass (FFM).30

Complaints were assessed in nine studies with the modified Medical Research Council dyspnoea questionnaire (mMRC),26,28,44 the COPD assessment test (CAT),35,37-39 the clinical COPD questionnaire (CCQ)35,36 and the functional assessment of chronic illness therapy-fatigue (FACIT-F).24

Functional impairment was assessed in five studies via physical activity levels.24,25,27,40,42 Physical activity was objectively assessed by daily step count25,40 and the time spent in light24 or higher intensity (i.e., >3000 vector magnitude units (VMU))42 activities. The time spent indoors/outdoors25,27 was quantified based on a diary.

Finally, QoL was assessed in twelve studies with the Saint George's respiratory questionnaire (SGRQ),27-29,31-34,43-45 the chronic respiratory disease questionnaire (CRQ),28,34 the 36-item short form health survey (SF-36),35,43 the 12-item short form health survey (SF-12),33 the EuroQoL 5-dimension questionnaire (EQ-5D)35 and the centre for epidemiologic studies depression scale (CES-D).41

Fig. 2 provides an overview of the outcome measures used to assess each health status subdomain. No study assessed all the domains that compose health status. Instead, the majority of studies25,30-34,36-43,45 (n = 15) focused on a single health status domain.

Figure 2.

Schematic overview of the outcome measures used to assess each health status subdomain. *, activities >3000 vector magnitude units; 6MWD, six-minute walking distance; CAT, COPD assessment test; CCQ, clinical COPD questionnaire; CES-D, centre for epidemiological studies depression scale; CRQ, chronic respiratory disease questionnaire; EQ-5D, EuroQoL 5-dimension questionnaire; FACIT-F, functional assessment of chronic illness therapy-fatigue; mMRC, modified Medical Research Council dyspnoea questionnaire; QMVC, quadriceps maximum voluntary contraction; SF-12, 12-item short form health survey; SF-36, 36-item short form health survey; SGRQ, Saint George's respiratory questionnaire.

(0.18MB).

Two studies24,26 reported a significant reduction in the 6MWD in the first two to three days after the onset of an AECOPD, which was found to be more pronounced in patients with severe to very severe disease (GOLD stages 3 and 4).24 In addition, studies exploring the effects of repeated AECOPD26,28,29 also found a consistent decline in the 6MWD however, the recovery trajectory was conflicting. One study24 reported that the 6MWD increased back to the pre-exacerbation status after seven days, whereas more extended studies26,28 found that the observed decrease in 6MWD was maintained up to two years of follow-up, with inconsistent results regarding a lower 6MWD in frequent exacerbators (i.e., two or more AECOPD per year).26,28

QMVC was significantly reduced three and seven days after the onset of AECOPD symptoms.24 A decline in QMVC over a 1-year follow-up was also observed but with no correlation with having frequent exacerbations (i.e., two or more per year).30 Yet, frequent exacerbators had a more pronounced decline in FFM.30

A consistent worsening of complaints at the onset of an AECOPD was found across all outcome measures.24,26,36,37

There was an increase (worsening) in the mMRC score within 48 h of the onset of an AECOPD, which was more pronounced in patients suffering from repeated exacerbations.26 Furthermore, this increase was sustained at 1- and 2-years follow-up in frequent exacerbators (i.e., two or more AECOPD per year), but not in single exacerbators.26 A second study also corroborated these findings by reporting higher mMRC scores in frequent exacerbators.28

Similarly, a significant increase (worsening) in the CAT total score was found at the onset of an AECOPD, which was already noticeable one day before and was sustained for up to seven days after the onset of the exacerbation.37 Consistently, a faster deterioration in the CAT score (yearly change) was found in patients with exacerbations in comparison to non-exacerbators.35 In addition, patients presenting a sustained worsening on CAT score (i.e., increase ≥2 points) had a higher number of exacerbations.38,39

An increase (worsening) in the CCQ total score was found at the onset of an AECOPD, which recovered in the post-AECOPD assessment.36 This worsening was consistent across all CCQ domains.36 The CCQ score also deteriorated over time but without differences between exacerbators and non-exacerbators.35

A significant reduction (worsening) of 5 points in the FACIT-F score was observed at the onset of an AECOPD, which recovered to a 2 points reduction at day 3 in comparison to stable phase.24

Daily step count was reduced during the first 7 days of an AECOPD compared to a stable week.25 Further analysis of these data showed that it took a median of 3.5 days return to baseline levels, and patients with the largest falls in daily step count during the exacerbation were the ones taking longer to recover.25 Conflicting results were found regarding the annual decline, with one study25 finding a significantly faster decline in frequent exacerbators (i.e., two or more AECOPD per year) in comparison to infrequent exacerbators, and another study40 finding no differences between the decline of steps in frequent and infrequent exacerbators.

Time spent in activities >3000 VMU was reduced during AECOPD in comparison to the preceding or subsequent weeks.42 This decline was significant during the first week of exacerbation, with no further decline on the second week, and tended to increase back to baseline levels in the two subsequent weeks.42 In line with these results, time spent in light activities was higher during the first week post-exacerbation than the second week, with frequent exacerbators (i.e., two or more AECOPD in the preceding year) presenting a more considerable reduction in time spent in light activities from week 1 to week 2 in comparison to infrequent exacerbators.24

There was an increase in time spent indoors during exacerbations, which was sustained in the post-exacerbation period (days 1 to 35 after the onset) in comparison to the stable phase.27 A decrease in outdoors time during AECOPD that continues in the post-exacerbation period was also found, with frequent exacerbators (i.e., at least 2.5 AECOPD per year) presenting a faster annual decline in daily time outdoors than infrequent exacerbators.27 However, inconsistent results were found as in another study,25 in patients with a similar severity of AECOPD but better baseline lung function, the fall in time and percentage of days outdoors during exacerbations did not reach statistical significance.

A consistent increase (worsening) in the SGRQ score with exacerbations27,29,34,43,45 was reported, with frequent exacerbators presenting faster and more pronounced declines in QoL.28,29,31,33,34,43-45 In a study with a 2-year follow-up, patients who suffered a single AECOPD improved in SGRQ score by −3.8 points while frequent exacerbators (i.e., two or more AECOPD per year) worsened +2.4 points.32

A similar pattern for QoL has also been found in the CRQ, SF-36 and SF-12. Patients with AECOPD presented a decline in CRQ, SF-36 and SF-12 scores, which was more pronounced in frequent exacerbators across all CRQ domains and in the physical component of SF-36 and SF-12.28,33-35,43 The EQ-5D was found to decline annually in exacerbators and non-exacerbators, similarly.35

A strong relationship between the number of AECOPD during the first year of follow-up and the change in depression score (i.e., CES-D) over 3-years has been found, with patients who suffered from more exacerbations presenting the most significant declines.41

A summary of the acute and long-term effects of AECOPD and the impact of single and repeated exacerbations on health status can be found in Fig. 3.

Figure 3.

Effects of acute exacerbations of chronic obstructive pulmonary disease (AECOPD) on patients’ health status. (a) acute and long-term effects of AECOPD on health status. (b) effects of single and repeated AECOPD on health status. ↑, increase; ↓, decrease; ↑↑, greater increase than in single AECOPD; ↓↓, greater decrease than in single AECOPD; *, only in repeated AECOPD; (?), conflicting results found.

(0.35MB).

An AECOPD is defined by an acute worsening of respiratory symptoms and it is usually characterized by increased airway inflammation, mucus production and air trapping.1 A recent proposal on an updated definition of AECOPD states that these events are characterized by dyspnoea and/or cough and sputum that worsens over up to 14 days, possibly accompanied by tachypnoea and/or tachycardia, and often associated with increased local and systemic inflammation.46 Not surprisingly, an acute worsening of complaints (i.e., dyspnoea and fatigue) during exacerbations was found.24,26 During this period, the worsening of airway obstruction results in increased work of breathing, dynamic hyperinflation and hypoxaemia, leading to symptoms of dyspnoea and fatigue.47-49 These symptoms usually recover following the exacerbation,47,49 yet persistent (up to 2 years) and higher increases in dyspnoea levels were found in frequent exacerbators,26,28 emphasizing the high impact of these events on patients’ life. In fact, the impact of the disease, as assessed by the CAT, has been found higher during either single or repeated exacerbations and on the long-term,35,37-39 and associated to the worsening of lung function, systemic inflammation and functional status.50,51

Exacerbations result in physiological impairment. There is an immediate reduction in QMVC three and seven days after the onset of exacerbation symptoms.24 The increased systemic inflammation appears to contribute to this, as QMVC has been correlated with systemic levels of IGF-I and CXCL8 in AECOPD52; and pro-inflammatory cytokines activate pathways leading to atrophy and inflammation-induced muscle dysfunction.53,54 The reduced physical activity levels and medication used (e.g., corticosteroids) during this period seem to also play a role in QMVC reduction.53,55 Moreover, the increased cost of ventilation during exacerbations increases resting energy expenditure, with negative consequences on body weight and muscle mass,55 highlighting a potential role of nutritional support in these patients.56 Over the 1-year follow-up, one study found that frequent exacerbations were associated with FFM decline,30 possibly indicating a cumulative effect of AECOPD on muscle mass depletion – the main constituent of FFM.57 Furthermore, the same study found that the decline in FFM was also associated with the use of maintenance oral corticosteroids,30 which has been described in steroid-induced myopathy.58,59 Given the known relationship between reduced muscle mass and muscle strength impairment,58,60,61 one could expect that frequent AECOPD would also be associated with QMVC decline. Nevertheless, no association was found,30 showing that muscle mass and muscle strength are not always reduced in the same proportion.62 Further research is therefore needed to enhance our understanding on the impact of AECOPD on muscle dysfunction.

Exacerbations (single or repeated) also result in an acute decrease in functional exercise tolerance, which was expected since the breathing load is acutely increased and patients experience breathlessness even when performing low-intensity activities.47,63 Moreover, fatigue and quadriceps muscle weakness also play a role as limiting factors of exercise performance.64 Surprisingly, although it is often assumed that AECOPD leads to a permanent impairment on exercise performance, it is still unclear if the decrease in exercise tolerance recovers after a few days,24 alongside with symptomatic recovery,6,47,65 or whether it is sustained on a long-term, with studies26,28 even suggesting a sustained impairment 2 years after the exacerbation. It is known that (i) different types of AECOPD result in distinct clinical findings, prognosis and responses to treatment66,67; (ii) hospitalized patients with AECOPD are the ones presenting worse prognosis68; and (iii) the use of antibiotics and corticosteroids presents inconsistent benefits depending on the clinical setting and severity of AECOPD.68 Since in one study24 only community managed AECOPD were included and patients were mostly treated with a combination of antibiotics and oral corticosteroids, while the other two studies26,28 involved a percentage of AECOPD that resulted in hospital admission and did not report the exact number of patients treated with each medications, it is likely that these factors may have contributed to the disparity of the findings. Moreover, the six-minute walk test presents a significant learning effect and evidence shows the necessity of conducting the test twice in AECOPD.69,70 Therefore, differences in the methodology regarding the frequency and timing of assessments and the number of tests performed, might have also contributed to the inconsistency in the results. Future studies with robust methodologies are needed to clarify the long-term effects of AECOPD on exercise tolerance.

Exacerbations lead to functional impairment observed by reduced physical activity levels,24,25,27,42 which seems to be more accentuated in repeated AECOPD. An acute decrease in physical activity levels is associated with the severe inactivity and low amount of time spent in weight-bearing activities during hospitalization for AECOPD, general immobility and tendency to become housebound.16,71,72 The worsening of symptoms - particularly dyspnoea at rest - hypoxaemia, muscle weakness and loss of exercise capacity might also reduce physical activity levels.71,73,74 In turn, reduced physical activity levels lead to further skeletal muscle deconditioning and reduction of exercise capacity, bringing patients into a vicious cycle of symptoms and inactivity.74-77 It has been hypothesized that this vicious cycle could explain the long-term effect of AECOPD on physical activity levels.74 Nevertheless, conflicting evidence was found,74,78 with studies showing that physical activity levels can either recover in a few days/weeks or may not return to pre-exacerbation levels, especially in the case of frequent exacerbators.25,27,40,42,79 The differences in the timepoints of assessment and outcome measures used (e.g., objective vs. subjective measures) might have contributed to the heterogeneity in the results found.80 Further studies, following the international recommendations on how to measure physical activity in patients with COPD,81 are needed to better understand the impact of AECOPD on physical activity over time.

The effects of exacerbations on QoL were the most studied. A long-term decline (up to 8 years) in QoL due to AECOPD was found,27,29,33,34,43,45 which was more pronounced in frequent exacerbators,28,29,31,33-35,43-45 suggesting a cumulative effect of repeated exacerbations in this domain. It is known that AECOPD have a huge impact on patients' everyday activities (e.g., walking, sleeping, work) and, consequently, they feel unable to maintain their lifestyle and make plans.82-84 Given these reasons and all the consequences of AECOPD mentioned above on complaints, physiological functioning and functional impairment, the decline on QoL was expected.

In sum, heterogeneity amongst the presentation and trajectory of recovery of exacerbations was found, and there is at least a subset of patients presenting a sustained worsening of health status after (repeated) exacerbations. This heterogeneity might have been influenced by the variety of AECOPD definitions found, as it is known that even small changes in the definition used affect the incidence rate, type and classification of exacerbations, with event-based AECOPD being usually considerably less identified and in specific groups of patients.23,85,86 The underlying cause of the AECOPD (e.g., viral exacerbations are known for being more severe and taking a longer recovery time),87 its severity, treatment setting (e.g., hospital vs. home), the standard of care provided (i.e., pharmacological treatment), presence of comorbidities, socioeconomic status and/or knowledge about the disease, might have also contributed to the heterogeneity found.67,87-93 A more accurate definition of AECOPD and understanding of its aetiology and diagnosis is, therefore, critical to better recognise exacerbations’ clinical impacts and improve treatment strategies. Despite this heterogeneity, AECOPD are usually treated uniformly (i.e., bronchodilators, systemic corticosteroids and/or antibiotics) without considering the different underlying outcomes and treatment needs.94 Since one size does not fit all, comprehensive health status assessments that allow the identification of distinct treatable traits amongst individuals are crucial to personalize treatments, contributing to improved AECOPD recovery and prevention.95 In this review, we have found that no study assessed all the domains that compose health status, thus future studies should explore the effects of AECOPD on patients’ health status using comprehensive measures. Moreover, early pulmonary rehabilitation is a safe intervention for the management of patients with AECOPD that has been shown to improve QoL, and reduce the length of hospitalization, hospital readmissions and mortality in these patients, while targeting several treatable traits (e.g., physical activity, exercise capacity, muscle weakness, dyspnoea) that are associated with exacerbation recurrence.4,96-98 Evidence suggests that pulmonary rehabilitation may be offered to patients with AECOPD to recover their pre-exacerbation health status.99 Pulmonary rehabilitation seems particularly important to those presenting a late recovery or who never recover to pre-exacerbation levels.90 Future studies should focus on personalizing pulmonary rehabilitation programs to target the different identified treatable traits during AECOPD.94,95

This scoping review has some strengths and limitations that need to be acknowledged. To our knowledge, this is the first review of the impact of AECOPD in the different sub-domains of health status in patients with COPD. A thorough search and screening were performed, and rigorous methodological and reporting frameworks (JBI and PRISMA-ScR) were followed. Nevertheless, we did not publish a protocol of the study before conducting this scoping review, thus the methods were not peer-reviewed prior to our search. Several concepts have been used to define health status and numerous tools were designed to assess the different aspects of this comprehensive measure. To ensure clarity, this work followed a previously published assessment framework for health status in patients with COPD.18 Additionally, the summary of the impacts of AECOPD on the different domains of health status was challenging due to the diversity in exacerbation definition and diagnosis, timepoints of assessment, outcome measures used, and lack of clarity of some results found. Most studies included a population combining different AECOPD severities, treatments and/or treatment settings, which prevented the presentation of these results separately. All these aspects have hampered results synthesis. Lastly, since most of the studies included were focused on moderate to severe AECOPD, the findings of this review cannot be translated to mild AECOPD.